air z . we > “ron vated senta fre eet re
Fur mr. ome Be IE
nee . x i henhehatehtgein seine
rei em ea own Semen ee = OU FRR a pavers on re whe aw hi re
nr - - : “ : - > ar Aa lassen piers eve es, 2 - a” p n were ae een os tevatrapyeoeey tre “ . va: : : De fist TiS pni pss rsrebre dps isbana Poa Asha obe re te! RE : - u : va tertile ad zur u vera tere MDs. Al trade : ET N a , years are er 2 sahen . a en aaa
ie ee
bate De poate
D
mr i ERE fore?
a an
TEN
Mill li li: Ve a TE ye ae i Fis Weg Be a6 NT ee ee ene ehe > me, | pA rad! “a wwe ified | Nn Står "INA ys ity “We a vv LM 1 Yan Ss NI tata J, te eIZ USL? sf Ad ha | (WN hs EL Sr vv | AN una wc a AN are rss BE vd
eee 74. ALP, ar 44 weht ver won os a KT dött frön ses NA P a”) rw hg Va | Eur a NA, LAN S = é "I TILA Ne we — *EL u ve Prey, | N Darrel hare co Mes yes n” \ % nun, Ni A re ly k || meyvv une well.» Fe alent FIVE TA ge ann MAN Yat 9 WA Put wre FT NAC AN I | : AMIENS 4 Min a ER © Array wie ee 7) Ar | SuM Aj ‘peal üb An
N X Meee ou ty) LT vor vorghh’ , a PETE Au % WV ~ ei, nie» ER ar a Mm Rp In... ov DOGO ne rae NPD SEA my a ' Å ats KAA Ad i | ae Agee Wed AO ol olelole lal S tt Ä ve LÅ NG 4 ? , gy>* a sans gal! Tt ul # Vat P45 -* Wer t + ex : AN AAR Ww tei ng GO Aad, we Bish ir 2 3 i m, Neon 4
Gee: aS Yue 5 we... HONG U hee dre ie. ver a a = odd Sat eA | wo & Rau s INN 3 + Vv Wr. nt gee USS arerk An Ways Vwi, ify
=v 6% Pink 4 44727 NN ww We aL Leaded MILK A 8 a b fr‘ es. os, es te a || | ESCHE RER AIR RR rede I
ANN det . Se ge, wT LT 049 ads INGA
pe var them
br ef “er, . wy > ( Ada > Ng We TN, LU MP II Ka
Yor Tha TITS VP een tt ult 7, LÄN ew e as a= Pur | Mee dt: zen Un
SNARES Spent a, Nin vv Pern. ON
IRA KIMMEL JA > wu lee | un“ ua A slet ar an Lad „RL
i; Aut inf TMA
"ver “N \ Met r-> > ah AK, TO Da, TITT Qs NI N, 1 eee SÖNER Were ie H RE,
U
PL 4 4 Nyt ef Yun, un
FRE NR AA ON, LAU
Whg A : SLI KIT 2 m
„nen + bl Zieh NR uw. in \ ee (Vv SANT CNS Ar or vatt a The vv |
En. VN Ayr rate fh rdr ee) aay That ¢ en ser” Sy , flyger A
u, vie Met ha N! RE SRA TOR oe a: Ni ly OH, nt Sivan mn... Il WHERE EEE IE PT ot fl 1 1 |. Some eign \ MS We" upg
Nye Oh neat N SET TEE HELLE 1 PVSIURN ee i ail Ten. vor er — a Ny Wp N "tive Se TALL UTN : | NUN Ih IS 4 " > RAA ML ik LJ Sn i | $\ MM Neuse RES, Aa wwe TD Ar | | eee UTNE NWS Yn
Ue ; | -T LIT NA,
RI wey SSR > ee UL wes HAA un: "teg RR RS
= 1. @ \ nz ER y A . aye ; 6, q bj y Md | N at 2 eA ah a MN
: vi AMI eh wenn” oc u. " | ee, 2 RRA AL
7 AN ha BA ALAR) \ ue ETT io rae Ed es F wy v. MALIN „ei Pe ie vy vu vr
ALL N er 1171 NETT uw PEPER | ae z yt" wy
ea ll Mena nanan
A UT LT “wre le.
irn ae
w’ Di Arv TI ANA
UNTEN
ws NRe- wa dt ohh Ta
< TUVA NS Pad ! =: en un Wan TN Y er ren \ eg u vo STIER en STEEL vå ND SN hee || | nates un NER te une, LU NULL SYN ISS um 4: "vy FETT Terre VÄN. IN | u. NN S5 2 tL EL TT LA tee ES ‘ Y ded ve Bi =i we wet age OS OT et NR RR 4 ee" 5 (My, nå wd vg ied, > Tt eld i S| a N \ sei N Wry og LE, y <3 Sv _ Wyse vetter? al NM SMA HT || vv ert u we: aney vy vv vv u yuviy a mwah q ie Vv ~ ae Vy ® i uN we q war NY JAA "Case oe Pinta MLE ALLA Lt 5 > ewehy CTT LT i Se wrens wo a sil Vv art u ee
IAN N aie u
ee
h : : i | : | un
A - 7 P } VR 1 3 ; x : aes } Aa i På se - A ? A ; : > Ng \ N sa da MN 6
aD 4 2 a
ey : | : ; | . p i i a
MATHEMATICA ET PHYSICA
PANT RN & "5 : ;
ACTA ACADEMIAE ABOENSIS
MATHEMATICA ET PHYSICA | |
ZEN 1) = \
ÄBO AKADEMI ÄBO 1922
INNEHÅLL
Nils Pipping: EIN KRITERIUM FUR DIE REELLEN AL- GEBRAISCHEN ZAHLEN AUF EINE DIREKTE VERALL- GEMEINERUNG DES EUKLIDISCHEN ALGORITHMUS BEGRÜNDET ni aa a Sk ean ES ER mail:
Helge G. Backlund: ON THE EASTERN PART OF THE ARCHENBASALTLPLATEAD N ee
Karl F. Lindman: ÜBER EINE DURCH EIN DREIDIMEN- SIONALES RESONATORENSYSTEM ERZEUGTE INTER- FERENZ DER ELEKTROMAGNETISCHEN WELLEN ...
H. Hausen: ON THE LITHOLOGY AND GEOLOGICAL STRUCTURE OF THE SIERRA DE UMANGO AREA, PROVINCE OF LA RIOJA, ARGENTINE REPUBLIC ....
Bruno Saxen: DIE LICHTEMISSION UNTER DER WIR- KUNG MOLELEKTRISCHER KRÄFTE AN DER OBER- BEACHES VON KRISTALERN Er a Hann,
Erik Hägglund, Nils Löfman, Eduard Färber: ÜBER AZETYLCELLULOSE AUS HOLZZELESTOFFEN ......
Nils Pipping: ÜBER EINE VERALLGEMEINERUNG DES EUKLIDISCHEN ALGORITHMUS ....... LEBEN
1— 16
1— 53 1— 29 1— 139 1— 16 1— 24
ACTA ACADEMIAE ABOENSIS MATHEMATICA ET PHYSICA I: ı
EIN KRITERIUM FÜR DIE REELLEN ALGEBRAISCHEN ZAHLEN
AUF EINE DIREKTE VERALLGEMEINERUNG DES EUKLIDISCHEN ALGORITHMUS GEGRÜNDET
VON
DR NIESCPIRPIN:G
Z. ZT. LEKTOR DER MATHEMATIK AN DER AKADEMIE ÄBO
ÅBO AKADEMI ÅBO 1921
EIN KRITERIUM FUR DIE REELLEN ALGEBRAISCHEN ZAHLEN, AUF EINE DIREKTE VERALLGEMEINERUNG DES EUKLIDISCHEN ALGORITHMUS GEGRUNDET.
1. Wir betrachten ein System von n+1, (n=1) positiven reellen
Werten eo Wo, VR (EO
Falls sie alle unter einander verschieden sind, bilden wir aus dem gegebenen System z neue Systeme nach der folgenden Regel.
Das u:te System (u=1,2,--,n) wird erhalten, wenn der grösste Wert, v,, durch den Wert U— Vy ersetzt wird, wahrend die Werte U,,U,°",Un Unverändert gelassen werden.
Die neuen Systeme bestehen dann sämtlich aus n+1 positiven - Werten; falls jedes derselben nur unter einander verschiedene Werte enthält, wiederholen wir dasselbe Verfahren in Bezug auf sie alle. Hierbei erhalten wir nm? neue Systeme, u.s. w. Bei dem »:ten Schritte werden a” neue Systeme erhalten.
So lange kein einziges von den betrachteten Systemen zwei gleiche Werte enthält, setzen wir das genannte Rekursionsverfahren fort; die bei dem folgenden Schritte hervorgehenden Systeme bestehen dann aus lauter positiven Werten. Falls wir aber ein System bekommen, das zwei gleiche Werte enthält, brechen wir das Verfahren ab; in diesem Falle sagen wir daher, dass der Algorithmus abbricht; sonst lässt er sich ins Unendliche fortsetzen.
Für z=]1 finden wir offenbar den Euklidischen Algorithmus wieder; nur ist jede Division durch ebensoviele Subtraktionen ersetzt worden, . wie die Quote angibt.
4 EIN KRITERIUM FUR DIE REELLEN ALGEBRAISCHEN ZAHLEN
2. Es sei w der Betrag einer reellen Grösse, die ihrem Zahlenwert nach gegeben ist; die Potenzen
Py ie n \ 1, w, w?, „®
sind dann n+1 bekannte positive reelle Werte. Wir wenden das oben angegebene Verfahren auf sie an und erhalten hierbei einen eindeutig definierten Algorithmus, den wir den zu w gehörigen all- gemeinen Algorithmus n:ter Ordnung nennen.
In Nrr. 3, 4 und 5 werden wir beweisen, dass der zu w gehörige allgemeine Algorithmus n:ter Ordnung stets und nur dann abbricht, wenn w eine algebraische Zahl höchstens vom n:ten Grade ist.
Diesem Satze gemäss besteht dann folgendes vollständige
Kriterium für die reellen algebraischen Zahlen n:ten Grades.
Eine reelle Grösse ist eine algebraische Zahl n:ten Grades (n>1) stets und nur, wenn der zum Betrag derselben gehörige allgemeine Algorithmus n:ter Ordnung abbricht, während der nämliche Algo- rithmus (n—1):ster Ordnung unendlich ausfällt.
Speziell ist die genannte Grösse eine algebraische Zahl \:sten Grades (d. h. eine rationale Zahl) stets und nur, wenn der zum Betrag derselben gehörige allgemeine Algorithmus 1:ster Ordnung (der Euklidische Algorithmus) abbricht.
3. Die bei unserem Rekursionsverfahren hervorgehenden Werte- systeme schreiben wir in folgender Weise auf.
Zuoberst, in die O-te Reihe, schreiben wir das Ausgangssystem ae (4=0,1,--,72). Allgemein tragen wir in -die »:te Reihe (v=1) die n” Wertesysteme ein, die aus den n” + Systemen der (v„—1):sten Reihe hergeleitet wurden, und hierbei schreiben wir zuerst die n aus dem ersten System der vorigen Reihe hergeleiteten Systeme; dann folgen die z Systeme, die aus dem zweiten System der vorigen Reihe hergeleitet wurden, u. s. w. Die a Systeme jeder derartigen Gruppe werden schliesslich nach dem in Nr. 1 festgestellten Prinzip geordnet.
EIN KRITERIUM FUR DIE REELLEN ALGEBRAISCHEN ZAHLEN 5 Wenn wir mit v så ) (v=0, ie 2; 2 n=1, Dur 13 n”)
das u:te System der »:ten Reihe bezeichnen, erhalten wir also das
Schema (0) S, (1) (1) (1) : Sa dre se a (2) (2) (2) (2) (2) (2) oe 2 SE RE Sea £13k
Jedem System sw gehören n+1 Werte, die wir mit u (4=0, 1,--, 7)
bezeichnen. Weil unsere Ausgangswerte w' sind, können wir
(0 (1) are @=0,1,--,n)
setzen, und allgemein: wenn wir aus einem gewissen System der (v—1):sten Reihe das System = herleiten, ersetzen wir den gröss- ten Wert jenes Systems durch einen neuen Wert nach der in Nr. 1 angegebenen Regel ohne die gegenseitige Reihenfolge der fraglichen Werte zu verändern; a bezeichnet dann einen eindeutig definier-
ten Wert. 2 5 Nach (1) lassen sich die Ausgangswerte u,,, in der Form
(0) N) 0 (2) 22 > Pir, 0” e=0 schreiben, wo
(0) RR SE. Pi. ön (=e)
(3) | N) ae = 0, (2 +0).
6 EIN KRITERIUM FÖR DIE REELLEN ALGEBRAISCHEN ZAHLEN
Falls wir sukzessiv mit sämtlichen Koeffizienten p dieselben Ope- rationen ausführen wie mit den entsprechenden Werten z, erhalten wir daher
Me v=0,1,2,*- | 5 0 IND (2) u 2 Doe 2 = n
7
in welcher Gleichung die Koeffizienten p ganze Zahlen sind. . Es sei AO die Determinante
(v) (2) (v) P 1,0,0 Pion Duo () (v) (v) (v) Az MO nl ann (v) (v) (v) m a,n,1 en
Wir haben dann Am A 1) ,(v=1), wenn wir mit 3 1) das Wertesystem bezeichnen, aus welchem das System 3 one
ist; denn AY geht aus A > hervor, indem man von den Ele-
menten einer gewissen Reihe dieser Determinante die Elemente einer anderen Reihe subtrahiert. Weil N ist [vgl. (3)], erhalten wir mithin
(5) AR =1, (v=0, 1,2, 50 = 1,2,.-,n?).
Aus dem Obigen geht nun unmittelbar der Satz hervor: Wenn der zu » gehörige allgemeine Algorithmus n:ter Ordnung abbricht, indem ein System S ” mit zwei gleichen Werten u a ZT
(2)
a erhalten wird, ist » eine algebraische Zahl höchstens vom n:ten Grade.
und u
EIN KRITERIUM FUR DIE REELLEN ALGEBRAISCHEN ZAHLEN 7
- Denn nach (4) haben wir unter der genannten Voraussetzung
3 (ri, Pia ajo’ =0
e=0
und nach (5) ist diese Gleichung keine Identität.
4. Umgekehrt nehmen wir nunmehr an, dass die ihrem Zahlen- wert nach gegebene reelle Grösse eine algebraische Zahl höchstens vom n.ten Grade ist, und wir wollen zeigen, dass der zum Betrag derselben gehörige allgemeine Algorithmus n:ter Ordnung abbricht.
Unserer Annahme gemäss genügt der genannte Betrag w einer Gleichung (6) | Bothy a4 +gn@"=0
mit ganzzahligen a Wenn wir &=U, si setzen, a sich diese Gleichung nach (1) in der Form
RR 20 0 (6y 5 0, 44-0
schreiben; wenigstens - einer von den Koeffizienten Ut a ist hierbei
positiv und wenigstens einer von denselben negativ, weil un et ist.
(v)
. Es gilt dann allgemein zwischen den Werten u, 3
se
eines beliebigen
Systems
(7) 2 U “ in =
eine Beziehung
mit ganzzahligen Koeffizienten, von denen wenigstens einer positiv und wenigstens einer negativ ist.
Dass dem so ist, wollen wir sofort vermittels vollständiger Induk- tion beweisen.
I
8 EIN KRITERIUM FÜR DIE REELLEN ALGEBRAISCHEN ZAHLEN
Nehmen wir also an, dass für ein gewisses System oe der »:ten Reihe die Beziehung
(0) 2 u nr
gilt, in welcher die Koeffizienten OR a die vorgeschriebenen Be-
bee
dingungen erfiillen. Wir betrachten das System Si der fo sten
Reihe, welches aus dem System ce in der fee hergeleitet wor- —
oy Or
den ist, dass der grösste von den Werten Hee sei es u,
or 2 durch den Wert De (A, +4,) ersetzt wurde. Wir haben dann
TD OO
(8) Mos A, My Ay Hy, åa (rl) 2 le) Sa Ans ETS nn : =F Ay),
und wenn wir
pete ees ; ur ); (9) ee i) (0) we My; A, u „= Un AAs)
wählen, erhalten wir nach (7)’, (8) und (9)
(7)" > Ue ee
Hos | A=0
Aus (9) geht hervor, dass n von den Koeffizienten der Beziehung (7)" mit n von den Koefiizienten der Beziehung (7) übereinstimmen, während nur einer von denselben neu ist. Weil ferner wenigstens zwei von den Koeffizienten der Beziehung (7)' von Null verschieden
EIN KRITERIUM FÜR DIE REELLEN ALGEBRAISCHEN ZAHLEN 9
ae sind, verschwinden also die Koeffizienten der Beziehung (7)” nicht
sämtlich; und weil alle Werte ae al 1) positiv sind (vgl. Nr. 1), folgt = ca 1)
‚daher, dass wenigstens einer von den Koeffizienten OM positiv und wenigstens einer von denselben negativ ist. Hiermit ist der folgende Hilfssatz gewonnen. Wenn w eine algebraische Zahl höchstens vom n:ten Grade ist,
eilt für jedes System S 2 zwischen den zu demselben gehörigen
Werten u“ eine Beziehung
(2) ©) xe „A
mit ganzzahligen Koeffizienten, von denen wenigstens einer positiv und wenigstens einer negativ ist. |
5. Als Antithese nehmen wir jetzt an, es sei der betrachtete allgemeine a on n:ter Ordnung unendlich.
Wir haben dann eine unendliche Reihe von Systemen sö) und
u®) Un, (1=0,1, -:, 7), welche die in unserem Hilfssatz angegebenen Be-
dingungen erfüllen. 0 Si Indem wir ein bestimmtes Koeffizientensystem UR 1 = 8, fixieren,
eine entsprechende Reihe von Systemen ganzer Zahlen
von welchem wir ausgehen, und ferner die Koeffizienten u ‚ (ez]) nach dem durch die Gleichungen (9) festgestellten Prinzip herleiten,
sind die Koeffizienten yy), sämtlich eindeutig definiert. Mit Hilfe derselben greifen wir für jeden Index » ein gewisses
von den Systemen s0) heraus. Wir bezeichnen es mit så) und in Analogie hiermit bezeichnen wir die Werte, die dem fraglichen
System gehören, mit a”, (4=0, 1,--,7) und die ‚entsprechenden
Koeffizienten mit ur ”) h
10 EIN KRITERIUM FUR DIE REELLEN ALGEBRAISCHEN ZAHLEN Das System SO ist mit dem System S, identisch; die Systeme
se) (v=1) definieren wir sukzessiv durch folgendes Rekursionsver- fahren.
Es sei das System se) fixiert. Für die Werte ul”), die demselben gehören, gilt die Gleichung
(10) > Uae
A=0
vr
und nach unserer Antithese sind hierbei die Werte ul”) sämtlich verschieden; der Grösse nach geordnet seien sie :
Die entsprechenden Koeffizienten der Gleichung (10) bezeichnen
wir mit vo), so dass diese Gleichung in der Form Ne (10) > VT A=0
geschrieben werden kann. Andererseits ordnen wir auch die Koeffizienten ul der Grösse
nach im algebraischen Sinne. Wir erhalten dann eine Koeffizienten- reihe, die wir mit
Ai Be,
0 n bezeichnen, und hierbei treffen wir die Verabredung, dass wir, falls UM) UV) und 2, <A, ist, in der Reihe (11) den Koeffizienten UL 1 2 1 links von dem Koeffizienten ay schreiben.
Jede Zahl w bezeichnet dann einen ganz bestimmten Koeffi:
IS DEN
~
EIN KRITERIUM FÜR DIE REELLEN ALGEBRAISCHEN ZAHLEN 11
(v)
4 zienten U. Falls der entsprechende Wert u mit w,” bezeichnet wird, nimmt die Gleichung (10) die neue Form
\
oy > w.” a HN 2=0
an, und hierbei haben wir
(1 2) (v) (v)
weil wenigstens einer von den Koeffizienten der Gleichung (10) positiv und wenigstens einer von denselben negativ ist. Nunmehr sind folgende drei Falle zu unterscheiden.
1:0. VÄ <0, Als das System se) wählen wir in diesem
Falle das System, welches aus dem System st) hervorgeht, indem
der grösste Wert, ol), durch den Wert N) ersetzt wird,
während die Werte al”), 2), >, 30 unverändert gelassen werden.
Die Werte al) und ol sind verschieden, weil vo, w”>0 ist.
2:0. ve >0. Als das System so wählen wir in diesem Falle das System, welches aus dem System se) hervorgeht, indem der grösste Wert, o), durch den Wert oe) wy”) ersetzt wird, während die Werte a”), al”), on unverändert gelassen werden. Die Werte w”) und 2) sind verschieden, weil v0, w<0 ist.
3:0. vr =0. Wir haben dann a 4g, al) +, und als
das System se) können wir folglich jedes beliebige von den beiden in den Fällen 1:0 und 2:0 genannten Systemen wählen Wir treffen die Verabredung, dass wir dasselbe System, wie im Falle 1:o wählen. —
12 | EIN KRITERIUM FÜR DIE REELLEN ALGEBRAISCHEN ZAHLEN
| Von den Koeffizienten ue? stimmen nach (9) wenigstens z mit rn von den Koeffizienten y) überein. Im Falle 1:0 wird statt es Koeffizienten we) ein neuer Koeffizient w vv erhalten, und im Falle 2:0 wird statt des Keenan we) der Koeifizient
we ve) erhalten, während im Falle 3:0 die Koeffizienten um sämtlich unverändert bleiben.
Wenn wir mit M () den grössten von den Beträgen u bezeichnen, haben wir mithin dem Vorigen gemäss
‚@=0,1, i ‚n)
wv) <m im Falle 1:0, (13)
w +) <M") im Falle 2:0,
und es ist daher jedenfalls |
(14) EO ENE
Die Grössen m” (v=0,1,2,--) sind hierbei positive ganze Zahlen, und daraus lässt sich unmittelbar folgern, dass es eine Zakl v, derart gibt, dass
(15) m) — mer) für jeden Index v=»,
ist. Denn sonst hätten wir nach (14) eine unendliche Reihe von abnehmenden positiven ganzen Zahlen, was widersinnig ist.
Wir können aber auch ein anderes Ergebnis betreffend Gleichheit oder Ungleichheit in den Beziehungen (14) herleiten.
Unter den Koeffizienten ut) seien a gleich MY») und 8 gleich — M (1) . Wenn dann u eine Zahl derart angibt, dass bei der Her-
leitung der Systeme Se) sea Soe ) der Fall 1:0
EIN KRITERIUM FÜR DIE REELLEN ALGEBRAISCHEN ZAHEEN 13
wenigstens «-mal und der Fall 2:0 wenigstens ß-mal eingetreten ist, haben wir nach (18)
(16) mi) sy +1) _ Eine Zahl a der genannten Art gibt es thatsächlich; denn wir können ohne Miihe folgendes beweisen.
a) Es gibt keine Zahl v' derart, dass ve =0 für jeden Index v=v' ist.
Es seien die Werte ul”) der Grösse nach geordnet
>...
Nach dem Bildungsgesetz des Algorithmus folgern wir mit Hilfe des Axioms von Archimedes, dass wir nach einer endlichen Anzahl
Operationen ein System erhalten, in dem der Wert ul”) der grösste ist, und nach noch einer endlichen Anzahl Operationen ergibt sich
(v')
ein System, in dem der Wert v, > der grösste ist, u. s. w. Früher oder später müssen wir also ein System erhalten, in dem ein be- U
liebiger von den Werten u“) der grösste ist.
Einer von den Koeffizienten ur) — sei es a — ist sicher von Null verschieden. Aus der Annahme vo für jeden Index v=v' folgern wir uw = u. (v= v’) und daher oe atu (ve). Wenn wir den Algorithmus so weit treiben, dass ein System sen (”=v') erhalten wird, in dem der Wert waa? der
grösste ist, haben wir also vr jus U 5 )+0 gegen unsere Annahme. 1
b) Es gibt keine Zahl v' derart, dass vi 20 fär jeden Index
v>v' ist.
Jedes Mal, wenn v) <0 ist, nimmt die Summe 2, von allen
positiven Koeffizienten ur ab, während diese Summe unverändert
14 EIN KRITERIUM FÜR DIE REELLEN ALGEBRAISCHEN ZAHLEN
bleibt, wenn vr =0 ist. Nach a) trifft der erstere Fall bei unend- lich vielen von den betrachteten Operationen ein, und wenn es einen Index der genannten Art gäbe, hätten wir also eine unendliche Reihe von abnehmenden positiven ganzen Zahlen 2,, was widersin- nig ist.
c) Es gibt keine Zahl v' derart, dass vy) =9 fiir jeden Index v>v ist.
Jedes Mal, wenn VO ist, nimmt die Summe ay von allen
negativen Koeilizienten ul im algebraischen Sinne zu, während
diese Summe unverändert bleibt, wenn V0 ist. Nach a) trifft der erstere Fall bei unendlich vielen von den betrachteten Operatio- nen ein, und wenn es einen Index der genannten Art gabe, hatten wir also eine unendliche Reihe von zunehmenden negativen ganzen Zahlen 3 was widersinnig ist. —
Nach b) und c) trifft bei der Herleitung der Systeme
ot alg Ds se + 2) nd
sowohl der Fall 1:0 als der Fall 2:0 unendlich oft ein, und es gibt daher eine Zahl wu, für welche die Ungleichung (16) giltig ist.
Die Beziehungen (15) und (16) lassen sich aber nicht vereinbaren. Ein Widerspruch wird mithin erhalten, wenn wir annehmen, dass der betrachtete Algorithmus unendlich ausfallt, und er muss folglich nach einer endlichen Anzahl Operationen abbrechen.
Hiermit ist der in Nr. 2 formulierte Satz vollständig bewiesen, und die Richtigkeit unseres Kriteriums ftir die reellen algebraischen Zahlen n:ten Grades ist daher dargelegt worden. |
6. Eine wesentlich einfachere Verallgemeinerung des Euklidischen Algorithmus als die im Vorhergehenden betrachtete wurde neulich von Viggo Brun! angegeben.
* En generalisation av kjedebroken | (avec un résumé en francais), Videnskaps- selskapets skrifter, I, Mat.-Nat. Klasse, 1919, N:o 6, Kristiania.
EIN KRITERIUM FÜR DIE REELLEN ALGEBRAISCHEN ZAHLEN 15
Aus dem Wertesystem v,5v,>-:>Un bildet Brun nur ein neues System, welches erhalten wird, wenn man den Wert v, durch den
Wert v,—v, ersetzt, während die Werte v,,V,,°:,Un unverändert
gelassen werden. Auf die Potenzen 1,w,--,w" angewendet gibt also der Brun-
sche Algorithmus die Systemreihe, die wir oben (vgl. Nr. 3) mit
a
i 0,1.2:--) bezeichnet haben, woraus hervorgeht, dass die Überlegungen in Nr. 3 ihre Gültigkeit beibehalten, auch falls wir statt des unsrigen allgemeinen Algorithmus den einfacheren Brun- schen betrachten. M. a. W. |
Wenn der Brunsche Algorithmus, auf die Potenzen 1,w, --,w" an- gewendet, abbricht, indem ein System so) mit zwei gleichen Werten erhalten wird, ist w eine algebraische Zahl höchstens vom n:ten Grade.
Wir wissen, dass der umgekehrte Satz für n=1 gültig ist, denn wie unser Algorithmus ist der Brunsche dann mit dem Euklidi- schen Algorithmus identisch. Es lässt sich ferner zeigen,! dass dem so ist ebenfalls für 2=2. In dem allgemeinen Falle aber, wo n eine beliebige positive ganze Zahl angibt, stellen sich Schwierig- keiten ein.
Versuchen wir nämlich denselben Weg einzuschlagen, der in Nr. 5 zum Ziele führte.
Nach unserem Hilfssatz (vgl. Nr. 4) gilt zwischen den Werten
des Systems st) die Beziehung ONO) > U, ROS 0 4=0 mit gewissen eindeutig definierten ganzzahligen Koeffizienten um)
es sei m“) der grösste von den Beträgen um)
* Auf diese Frage gehen wir nicht näher ein, weil Brun in einer noch nicht veröffentlichten Fortsetzung der oben zitierten Abhandlung ein damit äquivalentes Resultat erreicht haben dürfte. (Briefliche Mitteilung an den Verf.)
16 EIN KRITERIUM FÜR DIE REELLEN ALGEBRAISCHEN ZAHLEN
Um dem Gang der Überlegungen in Nr. 5 zu folgen hätten wir dann zu zeigen, dass die Beziehungen
(14 a) ea
gültig seien, und dass in denselben Ungleichheit unendlich oft eintrete. |
In dem speziellen Falle a=2 (und natürlich auch für 2 =1) lässt sich dieser Beweis thatsächlich führen; schon für 2=3 sind aber die Beziehungen (14a) nicht mehr gemeingültig, was aus folgendem Beispiel hervorgeht.
Eine Wurzel der Gleichung 2m?+2o%—2o—1=0 ist annähernd w = 0,8547; wir erhalten dann | | u =1,0000, u, =0,8547, ut, —0,7305, 1) —0,6243 Un 1,002 „U, 42,0, 02
und ferner u, =0,1453, u), —0,8547, u, =0,7305, u, =0,6248
(1) (1) (1) (1) Ufs UNSURE
Folglich haben wir Me 2, Ms und mithin M?< M,.
Es muss indessen ausdrücklich hervorgehoben werden, dass der betrachtete Algorithmus nichtsdestoweniger abbricht. Falls wir die Rechnungen ausführen, erhalten wir nämlich us = u —
Aus dem Obigen können wir keine bestimmten Schlüsse ziehen, ob man mit Hilfe des Brunschen Algorithmus eine Lösung des Problems von den arithmetischen Kriterien für die reellen algebrai- schen Zahlen n:ten Grades erhalten kann oder nicht. Wenn man aber den Brunschen Algorithmus durch den unsrigen allgemeineren Algorithmus ersetzt, ergibt sich — wie wir gesehen haben — eir derartiges Kriterium. !
Helsingfors, den 2. April 1920.
! Für geschichtliche Angaben über die frühere Behandlung des Problems verweisen wir auf die Vorlesungen über die Natur der Irrationalzahlen (Teubner, 1892) vor Paul Bachmann und auf die Dissertation des Verfassers: Zur Theorie der arith- metischen Kriterien für die reellen algebraischen Zahlen, Helsingfors, 1917.
ACTA ACADEMIAE ABOENSIS MATHEMATICA ET PHYSICA 1:2
ON THE EASTERN PART OF THE ARCTIC BASALT PLATEAU
BY
HELGE'G.!'BACKEUND
PH. D., PROFESSOR OF GEOLOGY AND MINERALOGY, ACADEMY OF ÄBO
ÄBO AKADEMI ÄBO 1921
En
CONTENTS
[CEU BAK GAIN CG) uC ee ee N pe le A d GEOEOGICAKLREATURESKIORITHE PEATEAUNGE RN. onion ee RS, NEW BVIDENCES OF JHE-ARCIIC: VOECANICVAGCTION 7. 22. 10 PETROGRAPHIC DESCRIPTION OF ROCKS AND ANALYSES A. BASALTS FROM THE FARTHEST EAST au BENNEIEEISEAND ca m en ae EN 17 DESAMLERRLESIIEISEAN DET Ay we ee 25 BENBASAEIEROM TEIESNEARESP EAST. ur 2 2 Sa EN 36 MODE-OFZEXTRUSION:OR-THIE/EASTERNIBASALTS. 22 04.5 2.22.22... 41 SON GCIENSIST ON SEIN EA Hr a N EN Rn ol
ON THE EASTERN PART OF THE ARCTIC BASALT PLATEAU.
1. INTRODUCTION.
In a recent memoir A. Holmes! has given an extensive petrogra- phical review of the basaltic rocks belonging to the Brito-Arctic province; he points out tnat descriptions of particular localities and their rocks are not lacking, and that specially the British region has become classic through the well known investigations by Judd, A. Geikie and A. Harker. He fills out the gap existing between other localities of the Arctic region previously described, by an in- vestigation, both petrographical and chemical, of a number of basaltic rocks chiefly from the Atlantic part of the Arctic plateau, sc. from West-Greenland (Hare Island), East-Greenland (Scoresby sound), Iceland, the Faroe Islands, Jan Mayen, Spitsbergen and Franz Joseph Land. The rock specimens described by Holmes were brought home by various expeditions at various times, but their position in space and time is not always strongly defined. In the classification of the arctic basaltic rocks on base of his investigations and on comparison with other recent descriptions of the same or neighbouring regions Holmes divides the basalts in saturated (quartziferous) ones and in undersaturated rocks, the latter comprising both olivine-basalts and nepheline-olivine-bearing rocks; the first group is repre- sented by all localities reviewed by this author except from Jan Mayen, the olivine basalt being the commonest volcanic rock of
1 A. Holmes, The basaltic rocks of the Arctic region. Mineralogical Magazine, vol. XVIII, pp. 180—223 (Aug. 1918).
6 ARCTIC BASALT PLATEAU
all arctic countries, whereon nepheline-bearing rocks were encoun- tered as trachydolerite on Hare-Island and Spitsbergen, ! as trachyan- desite and nepheline-basanite known from Jan Mayen and Spits- bergen. * Associated volcanic rocks are andesites (Greenland, Ice- land, East Spitsbergen), rhyolites (Iceland) and an alkali-trachyte (Jan Mayen). Throughout all rocks with a few exceptions are equal in salic and femic mineral constituents, and concerning alkalies distinctly sodic. On nine excellent and new analyses, made by H. F. Harwood, the author demonstrates some detailed peculiarities of the region, amongst them the belt of basaltic rocks rich in tita- nium dioxide and stretching from Hare Island and Scoresby Sound over Iceland to the Faroes; on both sides of this belt the richness in titanium dioxide fades out, the content in feldspar increases; this belt appears indebted to underlying magmatic sources high in titanium dioxide. In conclusion the memoir contains some remarks on ,petrographic provinces“, the connection of which with terms such as Atlantic and Pacific, alkaline and calc-alkaline Holmes finds not suitable. He attempts to mark out the petrographic series by one of their characteristic members, each series being limited in space and time by distinct geological processes. The andesitic, spilitic, trachydoleritic, and tephritic series are associated the first with compressional and upward movements, the second with com- pressional and downward movements in regions of sedimentation, the others with differential movements of faulted blocks, a dominant ' uplift or subsidence in connection with depth of fusion possibly having significance. To all these series basalts are common foun- dations, and basalts alone are not sufficient to determine the series. The Arctic region represents two series, the andesitic and the trachy- doleritic, as lying between two well marked types of petrographic provinces, the tephritic of the Atlantic Islands, and the andesitic of
1 These Spitsbergen rocks are described by V. M. Goldschmidt in Vidensk.- Selsk. i Kristiania Skrifter, math.-naturv. kl. 1911. N:o 9.
INTRODUCTION 7
the Pacific borders. Four sub-provinces are distinguishable in this region, namely the British area, the titaniferous belt, the area of Jan Mayen and younger rocks of northwestern Spitsbergen, and the area of older rocks of Spitsbergen, the basalts of Franz Joseph Land and those of the mouth the Yenisei river with the plateau basalts of the Yenisei-Tunguska region. — The dimensions of the fourth sub-province seem to be much too great compared with the others, and much too variegated by geological processes acting upon it during the volcanic aera of its development to be a simple one; this probable exaggeration in dimensions is caused perhaps by want of knowledge of details upon the North-Siberian plateau. The contributions to the knowledge of the basalts of the Yenisei- mouth, ! cited by Holmes, are too insignificant to conclude upon the whole eastern region, and the earlier descriptions of some Russian Authors (Polenov, Lavrsky), which embrace great parts of the Lower Tunguska- and Vilui-river basins, are too indefinite and too inanimate and the bulk analyses too incomplete to give an idea either of the character of these basaltic rocks or of their geological position. And finally, it may not be forgotten that the taimyrite, this remarkable nosean- and glassbearing trachytic rock, described by Khrustchoff,? probably is of late mezozoic or early tertiary age and belongs to the plateau; no analysis is given of this rock and no definite locality (,erratic from the Taimyr-Land‘) named, but an association with the basalts is more than sure.
A glance at the distribution of these basalts, given by Suess, ? begins from the south of the amphitheatre of Irkutsk, from Northern Mongolia and the Alps of Tunkinsk and then goes down the river
1 H. Backlund, Krist. Gest. v. d. Nordküste Sibiriens I. Mem. Ac. Sei. St. Pé- tersbourg 1910. XXI. N:o 6.
2 K. v. Khrustchoff, cited by H. Rosenbusch, Mikrosk. Physiogr. II. 2. 1908 p. 925.
3 S. Suess, La face de la terre, trad. par E. de Margerie, III. 2. pp. 942—956, peculiarly p. 948.
8 ARCTIC BASALT PLATEAU
Yenisei touching the Tunguska plateau and the river mouth, passes through the Atlantic region as far as the mouth of the river Mac- kenzie; then he mentions the basalts on the Siberian mainland | opposite the New Siberian Islands (Cape Svyatoy Noss) and returns to the point of departure over the Yenisei region. There are two gaps to fill out here: the great jump from the river Mackenzie to the New Siberian Islands, and the other from the mouth of the | Yenisei to Franz Joseph Land, the first being a very great one, | whilst the second is more insignicant and not much greater than | between Jan Mayen and Spitsbergen, but of an other character; | the space Jan Mayen-Spitsbergen is covered by the ocean and not | accessible to detailed investigations, whilst there are between the mouth of the Yenisei and Franz Joseph Land landmasses as Novaya Zemlia, the Solitude Island and the Nicholas Land, waiting for geo- | logical elucidation, the first of them in her eastern parts. |
The writer had the opportunity to hold in his hands an important | material of arctic plateau basalts from regions, which do cover the | whole ground of.the eastern area as marked out by Suess, and | hereto besides some material from the relieved interstices between the Atlantic and the eastern area at the east and the west, and the | petrological examination was advanced to the stadium which affords | the rock analyses, and some of them were already executed; but then the material was lost and the greater part of the field and | laboratory annotations were too, by accident lying outside the writer's power. Only small fragments were saved, and the author believes that it might be of interest to publish them at a moment, when the Scientific investigation seems to be awakened to new efforts.
2. GEOLOGICAL FEATURES OF THE PLATEAU.
From the investigations and conclusions by A. Geikie, ! Suess resumes that the whole plateau region is underlain by a series of great magmatic chambers communicating one with another and being independent in their shape of the mountain chains turned up in earlier time, and stretching through the region in various places. This statement gives some reason to expect basalt flows or rocks associated with them a little here and there, where crustal move- ments of the underground were favourable to the outburst of vol- canic action, however these movements had sufficient amplitude to set down to the very great depths which enclose the magma reser- voirs.. The difficulty to separate older volcanic rocks associated with the older underground ones, from rocks belonging to the plateau, arises immediately, particularly where the surface conditions or the - outburst power was unfavourable to the spreading of lava flows, or these were thrown away by a later erosion. The lack of mesozoic or younger sediments makes the task further difficult.
In the Atlantic region, wherever basalts belonging to this group are encountered, they are associated with upper mesozoic or early tertiary strata; even where the mainland is built up by palaeozoic or earlier rocks as in Eastern Greenland, a narrov coastal stripe of younger mesozoic or tertiary sediments nearly everywhere locates the geographical distribution of these basalts; on the backland basalt is wanted. The earlier basaltflows of Iceland are fixed as tertiary by intercalated fossil bearing beds, while the cone of Bee- renberg and the whole island of Jan Mayen seems to be of a more uniform cast without greater unconformities until recent time. The older Spitsbergen basalts (diabases) appear either as caps crowning the table mountains of mesozoic strata dissected by erosion, thus
1 A. Geikie, The tertiary basalt plateau of North Western Europe. Q. J. G. S. 52. 1896. pp. 331—406.
10 : ARCTIC BASALT PLATEAU
representing extrusive lava flows upon aj nearly perfect peneplain swept over the relatively undisturbed upper mesozoic sediments of a different age, consequently concealing smaller faults in the bed rock and developing slight unconformities at their base; to this group belong the basalts (diabases) of the north-western Stor- fiord, beginning from Mt. Teist and Mt. Edlund, culminating in the Mt. Hellwald group near Helis sound, and continuing on the shores of the southern Hinlopen Strait; the greater part of the Franz Joseph Land basalts is also reckoned hitherto. Or these basalts occur as homogeneous sills, simple and multiple, and as dikes of various sizes, as in the eastern part of the Storfiord region, on the shores of Sassen bay, at Cape Thordsen, in Bellsound (Recherche bay and others), and on King Charles Land at the east, always cutting through or enclosed by mesozoic rocks, and air exposed by denudation of different degree; they always seem to be in close alliance with faulting processes. |
Ås a matter of fact, there are two marks which always enable us to discover the tracks of the plateau, namely 1) traces of unfolded sedimentary rocks of upper mesozoic (or tertiary) age, and 2) the erosion base on these, or upon older even undisturbed rocks, the older the rock the greater in value the unconformity; this latter statement fixes these basalts as continental flows. Both these marks have their importance together (when tertiary rocks are absent) or each alone.
3. NEW EVIDENCES OF THE ARCTIC VOLCANIC ACTION.
A short review of the arctic coasts of the Asiatic continent at the far east, as far as they are known, gives us no strong evidences upon occurrences of arctic basalts in their close signification between Behring sound and Cape Svyatoy Noss. Washington describes foyaite and comendite rocks rich in potash, from Cape Deshnev, *
1 H. S. Washington, Igneous rocks from Eastern Siberia. Amer. Journ. of Sc. 13. 1902, p. 175.
EVIDENCES OF VOLCANIC ACTION 11
and Lindström gives a bulk analysis of a diabase-aphanite with 6.85 2/0 alcalies from Konyam bay.! Neither their age nor their geological position is made mention of. Mesozoic sediments, even in remnants, seem to be wanting there; the broad north going valleys and the flat topped rising ground between the different fiver systems seem to be an old and largely dissected peneplain with far driven maturity. Raised river beaches or old shore lines marked by basalt flows seem to be wanting. A preliminary revision of the rocks brought home by the 1909 coast survey expedition (Tolmachoff) could not discover any representative of this late group of volcanics. Even from Wrangel Land only older sedimentary rocks (probably of palaeozoic age) are brought home. ? |
The first clear evidences are from the Bennett Island group, in- cluding here the small islands in the southeast, as Jeanette and Henriette, and those more advanced to the east, discovered by Wil- kitski during the coast survey of 1915. The smaller ones of this group seem to consist entirely of basaltic volcanics and rise from the ocean level as steepsided cliffs. The geological structure of Bennett Island is a more variegated one, as shown by the deceased Baron Toll in his last report brought home with his collections two years after his tragic dead (1902). The oldest rock is a black grap- tolite slate of cambrian age in undisturbed horizontal position, expo- sed only in the deeper cuts of the island. Upon this floor, with great stratigraphical diastrophism, there rest marly continental depo- sits containing undeterminable fossil plant remains of jurassic age. The uppermost part of the sedimentary series is a soft yellowish
1 G. Lindström, Analyser af bergarter och bottenprof fran Ishafvet, Asiens nord- kust och Japan. Stockholm 1884.
2 On the map accompanying the „Explanatory text to the geological map of the part of Siberia investigated by order of the Siberian Railway Comittee“ (St. Peters- burg 1906, compiled by L. Jaczevski), there are indicated large areas of „andesite, partly basalt, trachyte and liparite“ on the shore and on the inland immediately west of Behrings sound; no exact report upon these mixed rocks is known to the
writer. — On the islands north of the Kolyma-mouth, according to reports of Wrangel there is some possibility to meet with young effusive rocks.
12 ARCTIC BASALT PLATEAU
sandstone of tertiary age separated by a denudation plan from the jurassic deposits and filling out the depressions and cavities carved out in these deposits by an earlier erosion. Between the two last mentioned sedimentary members of the series, following the plan of unconformity, one meets with a basalt sill of relatively great thickness, well exposed in steep cliffs in the middle part of the south-west coast, whilst the superimposed tertiary sandstone culmi- nates in the highest icecap-covered summit of the island. The sill seems to thin out in the northwestern and southeastern directions no basalts being reported from the hills forming the northwestern and southern capes of the island. Even from the northwestern shore, opposite the highest summit, and from the northern slopes of it no volcanics are reported, though deep erosional depressions seem to run through the island in a northeastern direction. The relations of the basalt to the roofing tertiary sandstone, if the con- tact is of igneous character or the sandstone of a transgressional type, are not cleared up; the former seems to be true by rate of the rapid outthinning of the basalt sill.
The New Siberian Islands are relatively well Lota through the investigations by the late Baron Toll and-his companions, but until to day no general report is published and not to be expected in the immediate future. Some facts placed to the writer’s disposal may be published here. The easternmost island, the true New Si- beria, is covered with quaternary deposits without any afflorements of older rocks. Ice drifted boulders and pebbles of basalt of the Bennett type are common upon the shores and are also found on the highest elevations of the flat island. The middle Thaddaeus island represents similar conditions, except on the northern promon- tory. On the western island, Bielkoff, Kotelnyi and on the cape Bereshnykh of the Thaddaeus Island, palaeozoic sediments of various. ages are encountered, the oldest of them being developed as camb- tian graptolite slates on the Bielkoff Island by the west. On the shores of the greatest island rocks of silurian, devonian and car-
EVIDENCES OF VOLCANIC ACTION 13
boniferous age are met with too, well characterized throughout by fossil fauna. The central part of the island, as tested in the upper parts of south running flat river valleys, is built up by mesozoic sediments; the Malakatyn hills covered by quaternary sediments are their culminating point. Only on few points beneath this cover tertiary sediments are preserved. The discovery of mesozoic strata of various ages explains the finding of ammonites by Hedenström during his stay on the island in the former part of the past century and whose report hereupon caused later on lively controversies. The whole region is strongly faulted, partly by concentric step faults, the Nerpalakh bay on the westcoast of Kotelnyi representing a circular stepfault, and partly by radiating faults; this circumstance explains the fact, that palaeozoic rocks are encountered only along the shores, while the mesozoic strata, also strongly faulted and turned up by faulting, build up the central part of the island. In this central part dikes of a glassy sodaliparite cut through the mesozoic strata forming low smothed hills in some river valleys. The relations of this volcanic to the tertiary strata are unknown. The rock is light gray, without coloured constituents, containing phenocrysts of quartz and twinned plagioclase feldspar in the glassy groundmass, and in steam cavities infiltrations of secondary silica (opal and chalcedony) are abundant. In the neighbourhood of these dikes an erratic boulder of a finegrained bright reddish granite rich in quartz was met with. Erratic pebbles of this sodaliparite were found also on the islands of New Siberia and Thaddaeus.
The basalts of the opposite mainland are mentioned by Suess; they cut through a gray granite and represent a magmabasalt type, tich in brown glass, phenocrysts of olivine and pyroxene, the feld- spars almost quite wanting. Pebbles of basalt are represented from the greater Liakhoff Island.
The continental backland is covered with quaternary and modern fine grained river deposits, the next step of connection being the north slope of the Kharaulakh range on the point where this range
14 ARCTIC BASALT PLATEAU
approaches the sea shore between the mouths of the Yana (Ust- yansk — Kazachyé) and the Lena (Tiksibay) rivers. From this point there are reported! flows (?) of columnar basalt and hot springs con- nected with them; no advice upon allied sedimentary rocks is given. From the river Lena crossing the Werkhoyansk range near Bulun no geological news are reported. The geological position of the basalt-(diabase-)sill (?) near Bolkalakh at the mouth of the Olenek amidst sediments of mesozoic age belonging to the mentioned range, is not further elucidated. Upstreams the river Olenek the nearly unbroken basalt plateau joins, extending to the upper Vilui river and its affluents on the one side, to the Anabar and Khatanga rivers on the other side, connecting with the Tunguska plateau.and, passing the upper Piasina river, with the basalt of the Yenisei mouth. A reference to the geological individualities of this Siberian basalt plateau in a narrower sense will be given in a further chapter. From the Alexei Island, between cape Tchelyuskin and the new discovered Nicholas Land, there is brought home a numerous col- lection of ice drifted pebbles of a typical fine grained diabase (ba- salt); the whole island is covered with quaternary deposits. By soundings there are drawn up from the sea bottom eastward from the island and southward, in the Thaddaeus bay, several pebbles of a similar diabase, covered with a manganese core. From the Ni- cholas Land the examined phyllitic slates, folded and contorted, are a proof of the rugged mountain chain reported from there. Greenstone samples brought home from the shore eastward of the Piasina mouth and from the islands off side the recently disco- vered Minin bay seem to belong, in spite of their basaltic origin, to an older formation; they are strongly crushed and metamorphosed. On the other hand, the basalt flows of the upper Piasina river, on the Noril Lakes, from where they are reported by Schmidt and others, belong to the plateau and follow the river valley down at
1 By the last wintering party (1902/1903) of the Baron Toll expedition.
EVIDENCES OF VOLCANIC ACTION 15
least to the Avam (Dudypta) river confluence, if not farther on; it is a typical surface flow, or several ones, as shown on the shores of the Piasina lake. Along the river downstreams they are markedly bound to the river beaches and the broader old valley has grown young again by the filling lavaflows, the present water wrinkle being compressed by nearly perpendicular cuts through the basalt ‘bed. The connection of these basalts to the Tunguska river plateau passing over the upper Khantaika, Kureika, Kheta and Kotui rivers seems to be uninterrupted. Pebbles of upper silurian fossiliferous limestone on the Piasina lakeshore prove the old age of the floor rocks. The coal measures of the Noril lakes and their metamorphic aequivalents farther south will be made‘mention of in a further chapter, and some explanation is also be given of the great exten- sion of the mesozoic area as presented on the compilative geological map by Ahlburg. ?
The diabases of the Yenisei mouth, specially these of the Kuzkin Island, were described earlier by the writer;? no new records upon them are reported.
The plant remains of upper mesozoic age from the Solitude Island are strongly silicified, but the short visit of captain Sverdrups expe- dition of 1915 revealed no volcanic rocks as were to have been expected here. ®
Erratic pebbles of basalt (diabase) from the Pancratieff Island on the northwest coast of the northern island of Novaia Zemlia were already mentioned by the writer. >? Since that time there were from this island brought home specimens of diabase (basalt) from the
1 Joh. Ahlburg, Die neueren Fortschritte in der Erforschung der Goldlagerstatten Sibiriens. Zeitschrift für prakt. Geologie 1913, p. 105.
2 Holmes (l. c. p. 221) supposes that the Kuzkin Island is identical with the Sibiriakoff Island; on older Russian maps the Kuzkin Island lies nearer the shore and is outfitted with the excellent harbour Port Dickson, meanwhile the Sibiriakoff Island is situated farther westward.
3 H. Backlund, On fossil plants from Solitude Island. Geol. Féren. i Stockholm Förh. 38. 1916, p. 265.
16 ARCTIC BASALT PLATEAU
east coast on the same latitude. The basalt crowns undisturbed sediments of mesozoic age, and seems also to occur as concordant dikes in this formation, the last strongly contrasting against the palaeozoic folded strata on the west-coast and representing by this means some ressemblance to the geological features of West-Spits- bergen. The relations of the geological structure of the east- and west-coast are not revealed on account of the central covering ice cap.
Åt last Franz Joseph Land by Suess is referred to the-Petchora _plain,! the former being the supposed extreme northern continuation of the latter. Already Count Keyserling, during his travels in the Pet- chora Land in 1843, encountered glacial drifted boulders containing marine fossils of upper jurassic age in the Bolshezemelskaia tundra. Later on, the original localities of jurassic rocks were stated on the southern shore of the river Petchora near the Ishma confluence. In 1909 jurassic red clay deposits, containing Cardioceras alternans v. Buch and Aucella Bronni Lahus. as characteristic fossils deter- mining their age as sequanian, were reported from the river Adzva, about 100 kilometres north of the confluence with the river Ussa, thus from a locality in the central part of the tundra. Only small remnants of these sediments preserved from glacial erosion are reported. Immediately in the neighbourhood of these deposits several small outcrops of a dissected basalt surface-flow, periectly glacial-smoothed, are met with on the river Adzva; the hills are partly covered with glacial deposits, but the geological relations of the basalt to the older sediments developed on the river, vic. to the carboniferous southward (Productus-limestone), to the permo-carboniferous (sandstone with plant remains) and, as mentioned before, to the upper jurassic clay in the nearest vicinity are not closely examined, unless they cut the former two formations. However, fragments of a sedimentary rock having great ressemblance to the jurassic clay and enclosed in the
Saye EE HT ICH CT , 1 E. Suess, I. c. p. 945: ,— — — ces iles affectent une structure tabulaire; elles sont situées entre les branches des Ouralides, et il est permis de considerer
le plateau comme un prolongement de la plaine de Petchora — — —“.
DESCRIPTION OF ROCKS AND ANALYSES 17
basalt flow suggest an upper mesozoic or even tertiary age. Even some evidence, that a strong denudation of sedimentary rocks has swept over the country before the basalt outbreak, is given by the _transgressional position on the permo-carboniferous met with in the few points where the floor rock was accessible to observations. This confirms the statement above that the outbreak took place in — continental conditions. Another proof of the youth of the basalt is given by a spring of constantly elevated temperature (earlier determinations —- 32°C, later ones +-28°C), which penetrates the Productus-limestone only a few kilometres further south of the Pymva-you river, left hand affluent of the Adzva; the spring is known since 1828 and tests the postvolcanic action by abundant sinter deposits. Although there is no strong evidence of the late mesozoic or early tertiary age of this basalt, and consequently no certainty of the rock belonging to the arctic plateau basalt, the characteristics are quite significant and uniform.
The basalts (diabases) of the Timan hills further westward are reported as being of palaeozoic age; they are rich in zeolite nodules. The basalts (diabases) of the Kanin peninsula, as described by Ramsay,! also seem to be of palaeozoic age, as the copper- bearing melaphyres and augitporphyrites too of the Novaia Zemlia’s southern island.
Herewith the ring around the pole is closed; some closer cha- racteristics of the eastern basalt representatives of the Arctic plateau are to be given in the following chapter.
4. PETROGRAPHICAL DESCRIPTION OF ROCKS AND ANALYSES. - A. Basalts from the farthest east.
a. Bennet island. The basalt from this island is, as far as can be judged from hand-specimens, medium grained and rich in zeolite patches and cavity fillings, the zeolites partly being so intimately
1W.Ramsay, Beiträge zur Geologie der Halbinsel Kanin. Fennia 32(1911—1912) N:o7.
18 | ARCTIC BASALT PLATEAU
grown together with the dark brownish rock, that it is impossible to separate them from the groundmass; the transition between rock and zeolites themselves by intergrowth is completely gradual and
in places projecting only as indistinct spots instead of unsharp —
phenocrysts in a normal porphyritic rock; there is no part of the rock without these spots.
The true amygdale minerals are represented by icositetrahedrons oi analcite, measuring until 3 cm in diameter; the milky white crystals are sharp faced and seldom pink cloudy dulled. Optically they are anomalous, with biaxial sectors of low birefracting power alternating with isotropic patches and stripes. —
On radiating, slender shaped prismatic crystals forming imperfect sphaerocrystals, the following optical properties were determined:
elongation positive, plane of axes lying in the zone of prisma,
y— acute bissectrix, thus „=; on flat shaped splinters, quite normal to £, in immersion liquid was found « = 1.18 0.001 y = 1.489 + 0.001. On thin section by measuring with Babinets compensator: y—a = 0.01:[74], thickness 0.037 mm. 7—B = 0.ooslo4] 3 Oro On free splinter with Babinets compensator: y—o = 0.008[35] bea i sitar? mae thus B= 1.481 and, @— «= 0.003; from the refractive indexes calculated: 2Vy = 63° [18] Dio 283.
All these constants agree well with natrolite. —
Another group of radiating coarse crystals with bright cleavage- surfaces and pearly lustre on them is intimately bound to the natrolite radiatings forming their immediate ulterior growth; the individuals are flat shaped, nearly leaved according to the best
DESCRIPTION OF ROCKS AND ANALYSES 19
developed cleavage-plane. Cleavage-flakes give good biaxial figures, with central emergence of the acute bissectrix y, and the plane of optical axes is normal to the elongation, consequently £=c with nearly straight extinction. The other cleavage-direction, neaily of equal perfectness and perpendicular to the just mentioned, gives the central emergence of the obtuse bissectrix «. Thus the optical orientation can be noted as nearly:
a=«e Dy, c=
Further on the following measurements in daylight were executed on the mineral with immersion liquid on free splinters of good orientation:
a = 1.525 + 0.001
= ‚| @=1.527-++0.001 | from these terms calculated: 6 = 1.527 + 0.0005 Na | Fa dl 2)
tel y = 1.539 + 0.0005 2 Ey = 82° 34’;
with the compensator of Babinet on free splinter: | 7: 8—a= 0.0032, thickness = 0.1326 mm.; on thin sections: ees 7— 8 0.0113, i — 754) | B:y—0 = 0.0145, 5 — (ORD ae Pees which measurements agree well with the birefraction values calculated from the refractive indexes. By means of the Wright micrometric ocular-scale there was found
2Ey= 81°40’, and a control measurement with camera lucida and rotation disc gave DE 82: ?
in good agreement with the above calculation.
A closer examination of thin sections developes a feeble striation quite normal to c, the sections normal to y and « being broad vic. narrow columnar shaped, and normal to £ elongated rectangular, and the extinctions sometimes sensibly deviating from the crystallo- graphical axes assumed as orthorhombic by the cleavages / (010)
20 ARCTIC BASALT PLATEAU
— perfect and / (100) — less perfect, and by the striation / (001). The deviation attains until 8”.3 on (010) and 4°—5° on (100) [c:8], the mineral being sometimes triclinic, but nearly and mostly orthorhombic.
Microchemical tests detected abundant lime, but also soda.
The optical properties indicate a mineral of the thomsonite group, the birefraction being too low and the refractive indexes too high as compared with the dates of Michel-Lévy and Lacroix,! the axial angle corresponding better to the values of Des Cloizeaux,? found on thomsonite from various localities. A series of controlling de- terminations of the birefraction of thomsonite made by Césaro? on thomsonite from nearly the same localities as that served to Des Cloi- zeaux gave lower values lying between 0.0026 and 0.0042 for B—e
a 0.008 „ Ooi „ vB
2 0.0138 „ Oo , y=e, and of the axial angle 2Ey—=82°—85°, and agrees better with the terms above. Prior? states that the double refraction of thomsonite sinks below the double refraction of stilbite (desmin), whose bire- fraction as given by Michel-Lévy and Lacroix® attains 0.006 (7 — 2) on columnar sections, which sections on thomsonite give the bire- fraction varying between s—a and y—ß, as given above; these terms rise to 0.oo6 (d—«) and 0.022 (y—Pß), conformable to the French authors. These discrepancies cannot be explained only by varying content of soda and lime in the specimens examined optically, because the analyses corresponding to the thomsonites from localities examined by Césaro show only small variations in these constituents, the raised refractive indexes perhaps being a test of heightened lime content, but the lowered double refraction probably connecting with the physical conditions during or after the mineral growth; the in-
1 A. Michel-Lévy et A. Lacroix, Les minéraux des roches, p. 320.
2 Cited after Dana’s System, 6th ed., p. 607.
3 G. Césaro, Bulletin d. 1. Classe des Sciences de l’Acad. Roy. de Belgique, 1908, pp. 259— 257.
4 G. T. Prior, On Sphaerostilbite. Mineralogical Magazine /2. 1900. p. 27.
5L.c.p. 319.
DESCRIPTION OF ROCKS AND ANALYSES 21
vestigations of Rinne! upon change of optical properties on zeolites by heating state lowering of the birefracting power (y—e) on thom- sonite. —
On the free ends the radiating groups of the thomsonite are often — covered with small radiating sheaves of a weathered yellowish mi- _mneral of a much stronger birefracting power on the cleavage flakes, but showing emergence of the optic normal (@) and an extinction angle (c:y) of about 18°, all that on the extremely few parts of the mineral, which are not changed into isotropic substance. The shape of the mineral and these few optical indications point to altered laumontite.
In the specimens brought home, no primary calcite nor silica minerals were detected amongst the amygdale and. vesicle fillings, whose order of deposition is: |
analcite,
natrolite,
thomsonite,
laumontite,
an order which denotes an increasing water and lime content during the gradual cooling of the rock in the late magmatic and postmag- matic (hydrothermal) stage of development, corresponding to a de- crease in soda, an inverted sequence as compared with the early and true magmatic stage. —
In thin sections the rock exposes an intersertal structure with phenocrysts of plagioclase feldspar and olivine only slightly exceeding the groundmass minerals by dimensions. The phenocrysts of plagio- clase are slightly twinned and distinctly zoned with a core corres- ponding to Ab,,An,, and an envelop of 45 °/o An; the outer contour of the plagioclase grasps in the groundmass by slight dentification. The lath-shaped groundmass-plagioclases are simple albite-twins with uneven border and an anorthite-content of 40--45 °/o and lower;
1 F. Rinne, Sitz.-Ber. Berliner Akademie 46. 1890. p. 1174.
22 ARCTIC BASALT PLATEAU
they are partly changed to thomsonite. The mafic constituents pre- dominate, but they are much decomposed: the olivine phenocrysts are nearly almost replaced by a homogeneous mineral with strong pleochroism (pale yellow to greenish brown) and high double re- fraction, biaxial negative with medium axial angle and a distinct cleavage normal to the acute bissectrix. These pseudomorphs are closely suggestive of the produce received by Thaddéeff! on heating olivine rich in ferrous silicate. The few unaltered remnants of fresh olivine allowed no definite determination of the ferrous silicate con- tent by means of refractive indices. The pale brown augite forms by dimensions the proportional transition to the groundmass; it is much altered by late magmatic influence, partly to chlorite and partly to a mineral of the serpentine group, and builds also up glomerophyric aggregates, intermixed with magnetite grains and ilmenite needles. Apatite needles are seen chiefly as inclusions in the plagioclase phenocrysts, but also in residual interstices. A strange ' appearence offer the great pseudophenocrysts of thomsonite corres- ponding to the zeolite patches mentioned above: they are surrounded by a narrow envelop of analcite and small quantities of natrolite needles, and although formed later than and at the expense of the enveloping minerals, as shown on sections normal to the zone of | prisma, they imitate closely veritable phenocrysts. — The small residual interstices are filled out by small quantities of analcite, and locally also natrolite needles are seen, but their quantity is unim- portant. No strong evidences of glassy residue were detected.
Since the mineral composition is being altered (hydrothermally?) and no rock fragment free from zeolite patches for analytical in- vestigation is obtainable, the measurement of the actual mineral composition by the Rosiwal method was not undertaken; never- theless an analysis of a specimen poor in zeolite was executed by Dr. N. Sahlbom, and gave the following results:
1 K. Thaddéeff, Die Olivingruppe. Zeitschr. f. Krystallographic 26 1896. p. 60.
=
Er Ans
er £0}
DESCRIPTION OF ROCKS AND ANALYSES 23
1 A B Molecular ili Figures en, Ba Mineral composition („Norm“) of Osann ers | cent. SiO, 44.43) 44.49] 45.08] 0.741 | 50.711Or ........... a 200.7 AI,O, 18.61) 13.05'|14.217) 0.183 | 12.101 Ab... ......... 19.32? F..56.40) A... 3.1 Fe,O, JAS. 2,78 ,.72.8810l022. 11, — Al. nn. 31 CHEN | FeO 6.62! 8.24| 8.59| 0.092 I 9.29|Cor .....-..-- 2.33 722200 | MgO 10.14 11.23) 7.01) 0.254 | 16.82 Sgn) 5803 st |W eee sas FEE CaO 6.78) 8.58} 10.20) 0.121 | 8.05 ; Le Aa) Na,O 2.29) 3.16] 3.99) 0.037) 2.45] * MesiO 13.47 CLES K,O 0.92 2.95) 1.79| 0.010 | 0.65,/7Y a uf 3.68 P..17.15) f...14.0 H,O at 105° | 1.58) 0.5 0.15] — | — Mg,SiO OS hk... 0.88 Oi 105° 12.4 0.05 — |. | ee Soe da 0..11.89 7... 0.95 Co, 0.02| 0.86] 2.42] 0.001 | — A TiO, 1:70] 2.69) 2.93) 0.021 | — ES RURGRN et P,O 0.33} 0.92] 0.62] 0.002 | — 275 VEN TEN NEE SL a An = OL ET a RE Behe S ; LAGE N | el 0.05 MnO 0.23) 0.17} 0.22) 0.004 | — SEN RFU FRAGEN NiO trace! 0.07;/nmone} — | — <'fem 37.80 BaO 002} — | 0.117) 0.0001) — H,O... 4.02 99.61) 99.71|100.41 100.00 100.55 Spec. gr. | 2.861) — | 3.01 Magmatic symbol: III, 5, 4, 4 | Auvergnose |
1. Olivine-trachydolerite, Bennet island. Dr. N. Sahlbom, analyst.
A. Olivine-trachydolerite, volcanic neck Mt. Halvdan, Spitsbergen. Prof. Dittrich, analyst. 4
B. Olivin-trachyandesite, Mary Muss bay, Jan Mayen. Dr. Har- wood, analyst. ?
Although the analysis differs conspicuously from those of the normal Arctic basalts in the low silica and high alumina content, the magmatic symbol is nearly accordant with the bulk of them,
1 Including 2.109/0 ZrO, and 0.0290 Cr,O,. 2 Including 0.05% V,O, and 0.01 9/0 Cr,O;. 3 Including 0.05%o SrO,
4 V. M. Goldschmidt, I. c.
5 A. Holmes, 1. c. p. 204.
24 : ARCTIC BASALT PLATEAU
thus giving a slightly erroneous idea of the true character of the rock. It was stated above that the mafic minerals predominate, the „norm“, on the contrary, states a slight predominance of salic constituents, since all lime and alumina are drawn to the salic group, though a great part of them is mafic. The figures of Osann enlisted above give a picture of more concordance with the true proportion, the great value of f being significant.
The classification of this rock presents some difficulty. Above analyses of two rocks irom the western Arctic basalt region are tabulated, but they differ in lower alumina content and in high alkalies, specially in potash. Moreover, in the Jan Mayen rock lime dominates over magnesia, thus the salic characteristic is more salient and the rock pointed out as alkaline by the normative (and actual) nephelite content; the same is valid to the Mt. Halvdan rock only on part of alkalies, all these marks being absent in the Bennet island-rock. Yet a comparison is made, because the high alumina does not conform with the demands of proper basalt.
It is seen, that whilst in the early magmatic stage of the rock the progressive crystallisation marked a decreasing lime content of the salic constituents (plagioclase), yet the late magmatic stage continuing in the postmagmatic creations marks an inverted deve- lopment with increasing lime content to the end of crystallisation, partly issued after the complete consolidation of the rock. If the inversion point, marked by the hiatus in cristallisation between plagioclase low in lime and analcite, is to be pushed backward to the earlier magmatic stage, or, in other words, the rock magma in early stage of crystallisation is to be supersaturated with a rising quantity of overheated steam, then the zeolite phase of the rock will begin before the groundmass is solidified (this moment being pushed forward) and a part of the minerals, specially these which are alkalibearing, will be bucked out and deposited in steam cavities, whilst the minerals richer in lime partly resist and partly alter in zeolite; this alteration in case of plagioclase —> thomsonite is ac-
DESCRIPTION OF ROCKS AND ANALYSES 25
companied by rising content of alumina, water, and partly lime, by lowering of the content of alkalies (soda) and silica and by lowering of the specific gravity of the rock. Only in this way the
7 early cristallisation of thomsonite is intelligible and the forming of
its pseudophenocrysts is drawn into true light. The real content of soda zeolites in the rock is very low, as showed by the analysis ‚figures, yet amongst the cavity mineral they predominate, and the relative soda content in each of them exceeds the soda content in the albite aione.
That something of this kind has worked out the actual stage of the rock, whose mineral composition is not due to the weathering process, finds a confirmation in the mode of olivine-alteration and perhaps in the lowered birefraction of the thomsonite mineral. Yet this process does not exclude an original rock of the calc-alkali series, if not the order of zeolite-separation might strengthen an alkali affinity. The olivine rich in ferrous silicate gives identical hints. The statement by Holmes,! that among the volatile fluxes of the alkali series carbon dioxide was a dominant member, cannot be strengthened in this case, no carbonate mineral having been found among the cavity fillings and only traces of them in the rock.
In consideration of this the writer resolved to rank the rock among the trachydolerites of the alkali series unless a hasty examination of both the rock and the analysis should rank her with the common basalts. Except alumina and alkali content and the lower percen- tage of titanium dioxide, perhaps individual, the analysis agrees well with the analysis of the olivine-trachydolerite of Mt. Halvdan.
b. Wilkitski-island. This little island east of the Jeannette and Henriette islands is represented by the material brought home by the Hydrographic Expedition of lieut. B. Wilkitski in 1913. There
1 A. Holmes, On the tertiary volcanic rocks of Mozambique. Q. J. G. S. 72. 1917. p. 253; this statement agrees quite well with the fact of increasing lime- content to the end of the postmagmatic stage, the present rock going not farther then to lime-silicate and not reaching the carbonate stage.
26 ARCTIC BASALT PLATEAU
are specimens representing partly a dense rock of darker or lighter steel gray coloured groundmass with sparingly dispersed phenocrysts of yellow olivine not exceeding 1 mm. in diameter; the air exposed surface of this rock is covered with a thin and smooth, rusty brown oxidation crust suggestive of desert varnish. Partly the specimens are slaggy and vesicular, representing lava flows, with rusty brown or glassy black surface and empty, no mineralized and smoothvalled vesicles, the rock being without visible phenocrysts. All specimens are of great freshness.
Under the microscope the darker dense rock reveals two genera- tions of all chief mineral constituents: olivine, augite, the lenads and perofskite, the feldspars quite wanting.
The phenocrysts of olivine reach the greatest dimensions, and the greatest among them show a zoned structure with stronger birefracting envelop and rounded crystal forms of hyalosideritic character. The axial angle is always negative. On a section | 7 of a well zoned phenocryst a measurement of refractive indices gave the following results:
in the core a ==1.670-+0.001, 8 =1.687 + 0.001 in the envelop =1.706+-0.001, = 1.31-+ 0.001; the double refraction measured with the compensator of Babinet upon a microphenocryst without zonar structure, thus corresponding to the envelop of the greater crystals, gave the following figure 7—0 = 0.0423, thickness = 0.022 mm. consequently y = 1.748 + 0.0015.
From the determination diagram of the olivine group, based upon the change of refractive indices in connection with the varying che- mical composition, * the core of the olivine is found to contain about 19 %/o Fe,SiO,, or 18 2/0 FeO, and the envelop about 36—37 %o of the Fe-silicate, or 32°/o FeO, thus the latter nearly according with hyalosiderite (34 %/o Fe,SiO,, or 30°/o FeO). Since the micropheno-
1H. Backlund, Uber die Olivingruppe. Trav. Mus. Geol. Acad. Imp. Sciences. St. Pétersbourg. 3. 1909. p. 77.
FE
DESCRIPTION OF ROCKS AND ANALYSES 21
; crysts strongly predominate, the medium olivine may be admitted
as of hyalosideritic composition. The core of olivine contains no
4 inclusions, only the outer envelop shows scarce rounded drops of
brown perofskite. — The second place in order of crystallisation
belongs to nephelite; only rarely he reaches the dimensions of true
phenocrysts, and then in intimate connection with olivine, forming
a parallel farther growth of idiomorphic olivine phenocrysts, as a
colourless likewise idiomorphic shell. The crystallographic axes c of both individuals are parallel, thus the axis of optical symmetry w (nephelite) is parallel to ® (olivine). Yet generally the dimen- sions rest below 0.01 X 0.03 mm, and in contrast to the groundmass- nephelite he is free of inclusions;! by this means it is possible to distinguish the nephelite microphenocrysts from the groundmass ones nearly equal in dimensions: the latter shows abundant central inclusions of apatite, augite, and sparingly magnetite. Both refrac- tive indexes of the nephelite phenocrysts rise above the refractive power of the balsam (n= 1.5354), while for the groundmass ones were fixed «> 1.5354, o< 1.5354. Consequently a zonar structure with a core of higher refracting power was proved on the phenocrysts; perhaps the core contains some quantities of RO-oxides (BaO?). — The microphenocrysts of magnetite are either built up by glo- merophyritic accumulations of several inviduals to a rounded grain, or they show sharp octahedrons. He is later im order of crystal- lisation than the greater part of the olivine microphenocrysts, be- cause in glomerophyritic accumulations of olivine, which are sug- gestive of „olivine-bombs“, he forms the cement together with ne- phelite and analcite. Diminishing in dimensions he passes into the groundmass-magnetite. No sure tests upon titanium-content of the magnetite were shown, but his near relations to perofskite makes such a one probable. — The augite-microphenocrysts are pale yel- low with brown-purplish tints. The core is free from magnetite in- clusions, but contains short needles of apatite and square sections of nephelite; an outer zone is overloaded with magnetite inclusions
28 ARCTIC BASALT PLATEAU
and corresponds in time to the initial growth of the first generation
of magnetite; the envelop is relatively free from magnetite inclusions
and corresponds to the crystallisation of the groundmass-augite. The contour of the augite is pointed out by a garland of magnetite, melting up together the dentification of the envelop. He forms elongated prisms (10 : 1), rarely flattened after (100), and then twinned; sometimes the prisms in their parallelism point out a fluctuation
structure. Extinction angles measured on sections strongly | £, gave: /
c:7=A47°.s in the distinct core, = 48°.5--53°.4 in the intermediate zone, gradual transition, = 50°.5 in the outer shell. The double refraction, measured with the compensator of Babinet, gave y—o = 0.023], thickness 0.030 mm., with n= 1.7. Axial angle: 2Ey—114°40’, in the intermediate zone TOR os Ir athevieGie 2 V7 — 109° 205 — (Ol Dispersion of axes: A—0o << v strong | B—o>v weaker| An irregural fissuring of the augite does not allow a measurement of the refractive indexes. The dispersion of extinction is considerable as tested on different interference tints on both sides of the extinction,
| calculated for 8 = 1.7
around 7
and conformably to the axial dispersion: ey, << :7,. The optical pro-
perties test an augite perhaps rich in titanium (dispersion!) and lime (? double refraction) content, containing soda too (extinction angle!).
The groundmass is very finely grained and inverted intersertal: the narrow and long prisms of augite build up a network, whose meshes contain the colourless constituents and partly needles of apatite, and whose nodes are represented by olivine individuals. In case of diminishing grain, which happens in „Schlieren“, the meshes lessen and their content too, the augite prisms growing thicker. The magnetite grains are confined along the augite prisms.
\ ie
DESCRIPTION OF ROCKS AND ANALYSES 29
The order of crystallisation is the same as pointed out for the phenocrysts, yet the nephelite-forming period closes after the growth of the greater part of augite. The last residue is a colourless cement with a refractive power much inferior to that of the nephelite and of.the balsam; sometimes there are seen tracks of cleavage and crystal forms indicating analcite. Belonging to the optical properties of the groundmass constituents, they agree well with those of the phenocrysts: the olivine always points out a hyalosideritic compo- sition, the augite shows hourglass structure, the nephelite is lower in refractive power. Among the nephelite sections there are some rounded of lower refractive power than the bulk of groundmass nephelite and higher than analcite, containing central inclusions of magnetite: the dimensions of the sections were too small to test if there really is leucite, yet the probability is close. — No traces of melilite were tested.
As accessory mineral great and sharp dodecahedrons of yellowish brown perofskite are characteristic; sometimes the crystal forms are wanted and it becomes drop formed, with a broad border of black magnetite, which is developed on the crystals too. The transition between core and border is gradual, by increasing intensity of tint. Cleavage cracks [(100) and (111)] are always well developed, the mineral is completely isotropic and is wanting inclusions. — No signs of weathering or secondary alteration are visible in the rock.
In the lighter rock, which in mineral composition agrees well with the just described one, the perofskite-mineral is more abundant and descends in the groundmass too, partly and to great extent replacing the magnetite. |
The slaggy and vesicular scoriaceous rock contains foreign in- clusions of pebbles and sedimentary rock fragments. The ground- mass in this rock is not differentiated showing dominating brown glass. In the dark coloured mass the few square and hexagonal sections of nephelite appear as windows. The olivine shows all transitions from yellow-brownish pleochroitic substance with the
30 ARCTIC BASALT PLATEAU
strong birefraction of the starting mineral to completely dull pseudo- morphs of iron oxide, due to late overheating of the outer parts of lava by escaping steam and gas masses during the eruption. — The vesicles are empty.
An incidental estimation by the Rosiwal method of the mineral com- position of the darker dense rock by volume is tabulated as follows:
Minerals: Volume %o: IAD ATER OC ian SE Za Maenente and peroiskites =: san. aes 6.0 Golonrless minerals. 2 en 44.5 Olivinerand.ausiter ke. oe ei ce 47.4 100.00
This rock was analysed by Dr.N.Sahlbom with the following results:
9 A Molecular Di Figures SHE Mineral composition („Norm“) of Osann bers | cent.
SiO, 41.18 | 44.66 | 0.686 | 45.72| Ne............ a L. 37.48 | 5 °° 45.7 Al,O, 169i 1297. ,0:115.) 7.23 |, Ze se ek EEE NEG ANSE PeO d.41 | 3.84 | 0.034). — Se a7 ke Cee 010 FeO 7.94 | 7.55 | 0.110 |f11.50 Pye 39g MgO 10.58 | 9.35 | 0.265 | 16.63 CaSiO, ... 8.00 Ih sia ty TE CaO oes 8.82 ou 10.87 | Di MgSiO, ... bal pois are Na,O 0.88 | 4.24 | 0.095 | 5.99 FesiO; se) ai C22 O20 K,O 8208: 15 72278: |. 0:033112.06.| Aine Ge sera N 6.01 I O9 H,O at 110” | 0.11 | 048] — | — R...:0.5 HO>110°| 0.17} 0.8| — | — Ol eo . . 14.70 M ..12.s1 co, none|none| —# — Besıo, 7 Saf O. .29.33 | M’.. 0.82 TiO, 2.34. 2:76 | 50.029')<° — | Am ne ee: 9.43 RIO 1.25 | 1.10 | 0.009 | — Cr,O, Day ee ee (ARN RN ee 08 MnO oe MESSER ee 4.87 NiO none) 0.26; — AR Aura. 27 A BaO 7 se ONE = fem 62.66
99.65 | 99.71 100.00 100.34
Spec. gr. | 3.045) — Magmatic symbol: TVG 2 OG 2 aaldose Ill, 9, 1, 4 Jivaarose1
1 The ratio sal: fem lies strictly on the limit between class III and IV; in both subdivisions of these classes nephelite basalts are richly represented.
Rn,
lyst.
DESCRIPTION OF ROCKS AND ANALYSES 31
2. Basaltoid nephelinite, Wilkitski island, Dr. N. Sahlbom, ana-
A. Nephelite basanite, Mt. Sverre, Spitsbergen, Prof. Dittrich, analyst. !
The deviation of this analysis from those of Arctic basalts as tabulated by Holmes is still more conspicuous than in the case of the Bennet land olivine-trachydolerite. With the young volcanics of Woodbay, Spitsbergen, described by Goldschmidt (I. c.), it agrees in magnesia content dominating over the lime one, thus being richer in mafic minerals; but none of them contains equal quantities of alcalies, particularly of potash, which in combination with low silica excludes every possibility of feldspar formation.
Even in this case the classification of the rock offers some diffi- culties; the important content of olivine and the low alumina points out a basaltic type, as strenghtened by the magmatic symbol, yet the intratelluric generation of nephelite together with the perofskite (and potash) content without melilite corroborate a nephelinite affi- nity, with leucititic dash. From all these rocks the described one differs with respect to the magnesia content exceeding the lime, having this mark common with the Mt. Sverre-basanite, from which it can be derived by raising the alkali content. The rock belongs more intimately to the alkali series than any of the Arctic basalts, retaining some of the pecularities of this group: it is an Arctic basaltoid nephelinite, for which the writer once proposed the name , onkilonite “|?
The mineral composition as given by the norm offers a close junction to the true one, differing only in details. The figures of Osann significate the rock as belonging to the scarce group, whose
1 V. M. Goldschmidt, |. c.
2 H. Backlund, Bull Acad. Imp. Sc. St. Pétersbourg 1915 p. 307. The early natives of the north-eastern coast of Siberia, near Behrings sound, called ,onkilon*, emigrated, conformable to traditions, under the press of new immigrants, to the northern islands in the Arctic ocean. Cf. F. v. Wrangel, Reise, Il, p. 220. Berlin 1839.
32 | ARCTIC BASALT PLATEAU
projection in the a—c—f-triangle lies on the a—f side. The actual composition („mode“) as calculated on base of the analysis and the © quantitative estimation given above is tabulated as follows, with the following modifications against the optical statements, namely that 1/ all TiO, is forming perofskite (it was supposed, that the magnetite contains small quantities of TiO,, and that the augite is titaniferous); 2/ that the leucite forms an independent generation in the ground- mass and that the composition of nephelite and leucite is an ideal one (no absolute certainty hereupon was given). The overplus of soda over the aequivalent of alumina, after nephelite and analcite (on base of H,O) calculated, was checked as acmite molecule be- longing to the augite; and the olivine: augite proportion was calcu- lated on base of resting FeO+-MgO-+MnO-content, i. e. the hyalosiderite composition of olivine and the remainder of SiQ,.
Mie AS Weight Volume percentage percentage
Olivine REN nace BE ARE NS 22.091 495 46.8 ISIE Ie Wren EEE RN 27.725 27.4 NepHelite ses... Cece ou Acts i Ue ee NE 21.50 25.0 NEUERE je ke rr ls occ bee alot be ro IRMA RSA ie 142] 37.85 ia} 45.2 ANT CEN ChE a SNC aya kilo shy EG he Oh a) Rae ems 2.08 2.9 IVIG GTI ULE ees RN RB, settee eee e eee et Ting re 5% Peroiskiteng a ren al RN IR 3.95 3.0 DAUER NE NE ae Aes MPSS EN a ON 2.73 2.6
99.55 99.6 SPEAHEIOTAVILYSHEL. Mur en SER NS LIGA 3 045 —
x ri GA LC een De ole 4 Va eeemionag — 3.042 |
A comparison of the figures found by measuring and calculated
gives a good concordance; a calculation of the lenads (except analcite) as extremely potash bearing nephelite, with a composition as for instance found in the sodalite-syenite of Kangerdluarsuk (Greenland) by Lorenzen,* or produced experimentally by Ch. and
1 A. Lorenzen, On minerals from sodalite-syenite, Greenland. Min. Mag. 5. 1883. p. 60.
DESCRIPTION OF ROCKS AND ANALYSES 33
G.Friedel,'! and for which kind of nephelite Morozevicz? proposed the formula K,Na,Al,Si,O,,, gives 32.4 °/o of it, but displaces the proportion of augite to olivine in favour of the former (45.48 °/o and 8.03 °/o), thus giving discrepancies both with the eyepiece estimation and between the specific gravity determined and calculated (lowering the latter). For calculation of specific Srey served the main values given by Rosenbusch.?
The figures of chemical composition of the augite calculated (sub 1, below) in the manner indicated above differ from a usual pyroxen composition in high lime content exceeding the sum of MgO-+ FeO, yet a comparison with augites from analogous rocks rich in lime and soda, as p. ex. the basalt of Rhön (anal. 3),* or in lime and potash, as p. ex. the madupite of Pilot Butte, Wyoming (anal. 2),° as tabulated below, shows some common features: —
| |
SiO, | 5220 | 50.86 AB | Al,O, we | 4.52 LURLSAON 7.9 | Tas | 6.54 | FeO | 6.06 | 1.82. | 3.26 MgO | 6.96 lf DE 12.01 | CaO I SOS sa Dal 101.81 | Na.O 3.00 0.76 | 0.80 | K,O = 0.42 | 0.36 lag OM SE! 0.31 0.11 | TiO, — I 3.03 1.44 | P,O, | Tras | war 0.36 | MnO | 0.54 | — -— | | 100.0 | 9916 | 100.35
Spec. gr. 3.4 | =| 3.402
1 Ch. et G. Friedel, Action des alcalis et des silicates alcalins sur le mica: production de la nepheline etc. Bull. Soc. Fr. Min. /3. 1890 p. 131.
2 J. Morozevicz, Über die chemische Zusammensetzung des Nephelins. Bull. Acad. Sc. Cracovie, Cl. Sc. math. et nat. Octobre 1907, p. 992.
3 H. Rosenbusch, Mikroskopische Physiographie I. 2. 4th. ed. 1904.
* X. Galkin, Chemische Untersuchung einiger Hornblenden u. Augite etc. N. Jahrb. für Min., Beil.-Bd. 29. 1910. p. 681.
5 Whitman Cross, Igneous rocks in Wyoming. Amer. Journ. Sc. 4. 1897. p. 130, W. F. Hillebrand, analyst.
34 ARCTIC BASALT PLATEAU
The two rocks analysed from these localities, whose geological history no doubt in time and in cause of volcanicity will be common, differ conspicuously in their mineralogical and chemical composition, the one being a typical alkaline rock, the other apart from the vesicle minerals showing a distinct character of the calc-alkali-series and approaching to a common basalt; only the high alumina content makes such a calc-alkalic affinity at least doubtful. To facilitate an immediate comparison the two analyses are tabulated together
as follows: l | 2 | | SiO, 44.43 | Alas .| | ALO, 18.61 11.69 | ‘FeO, | 3.48 541 | | FeO 6.62 7.94 MgO 101 | 1088 | CaO | 6.78 | 9.58 _Na,O Pee 8 | KO | 0.92 3.08 | H,O at 110° | 1.58 | 0.11 HO SS 110%) 1020 a a LIGOR | 0.2 | none | | TiO, eee oh irene | 11805 | 0.33 | 1.25 | MnO ot Ops tle tes One | BaO | 0.02 | 0.0 | 996 | 99:65 Spec. gr. 2.861 3.045 |
This contrast is a typical one for the basalts of the Arctic region, as stated by v. Wolff! and others, yet in no part, as known to the . writer, pronounced in such an extreme manner as in the case in question; the mixed character of the Arctic province was explained by its situation between the Atlantic and Pacific provinces too, being on both sides limited by them. In no part doubting the peculiar character of this province the writer wishes to direct the
1F. v. Wolff, Der Vulkanismus. I, pag. 152 (1914).
DESCRIPTION OF ROCKS AND ANALYSES 35
attention to the possibility of explaining the genesis of the two rocks differing in such a degree but closely connected one with another in space and probably in time too, by progress of cristalli- sation during the cooling process after the sheme given by Bowen for the silicate rocks in general." The content of mafic minerals,
being common and almost of the same kind in both rocks, except
small quantities of accessories, can be neglected in the following
discussion.
In decreasing the temperature the degree of dissociation of the magma-components diminishes and the cristallisation process can be derived from the ternary system: quartz (SiO,) — alumosilicate
of alkalies lowest in silica (i. e. Na(K)AISiO, [nephelite]) — a hypo-
thetical alumosilicate of lime lowest in silica (i. e. CaAl,SiO,).
Sid,
Alkalifeldspars RL
N Anorthite
Leucile-groun
NaVALSi 0, CoALSiO, IN ae l tle-group)
Kies.
Although the pecularities of this system are not yet known, the way of crystallisation of a solution, whose composition lies near the base of fig. 1 in x, can in some measure be foreseen. In an early stage of crystallisation the lenads will begin to separate, but after the bulk of mafic minerals crystallised out and the rest solution acidilied, the lenads become instable and due to resorption; the
I N. L. Bowen, in Journal of Geology, Suppl. to vol. 23. 1915.
36 ARCTIC BASALT PLATEAU
feldspars begin to separate following the decreasing lime content. If the magma is steam-overladen, the bulk of alkalies from the resorbed lenads can be redissolved during the uninterrupted epi- magmatic and postmagmatic flux-rising stage of the rock-develop- ment immediately pursuing the magmatic one; and then deposited in the steam cavities or elsewhere as amygdale minerals. In this manner the rock loses her primary or secondary alkalic character | and becomes an ordinary basalt, if the cooling was not going too fast and steam or gas was not wanting. If in exchange the cooling was going faster and the mineral separation was interrupted by quenching in some part y of the hypothetic curve figured as ap- pendix to x, the lenadic mineral composition may be fixed and consists of minerals in some regard instable.!
The two cases are realised by a rock body of moderate dimen- sions intruded on a gentle depth in a sedimentary series of at least some permeability to gases; and by a rock of surface-extrusion or in general of quick cooling. The basaltoid nephelinite shows all signs of surface extrusion, while the Bennet-land trachydolerite gives by her coarser grain and the local circumstances some signs of an intrusive sheet. These two examples do not imply an exceptional primary alkaline character of the magma, contrarily a slight one; but they give an explanation of the sporadic alkaline character of the Arctic basalts and this explanation can be extended to the | young volcanics of Spitsbergen.
B. Basalt from the nearest east.
The geological features of the volcanics of the Petchora plain are related in a former chapter (p. 16). The petrological character of this basalt is to be described.
1 The raising steam content to the end of crystallisation probably gives such a direction to the crystallisation way, and the true end of this curve is signified by particular redissolution of the feldspars, the latter being broken down and giving place to the zeolite formation as early as in epimagmatic stage.
ee ve
DESCRIPTION OF ROCKS AND Ae eRe 37
‘It is a dense rock, black in colour, only a few specimens repre- senting minute phenocrystals of plagioclase. The densest rock shows irregularly dispersed smooth valled cavities, empty on the air exposed rock surface, but containing a whole individual of cal-
cite only a few millimetres beneath the surface; and only one cen-
timetre or two deeper these calcite individuals are replaced by black sphaerules, which easily detach from the smooth walls of the cavity. Typical vesicular surface rocks are not represented.
The mineral composition is a simple one. The densest rock is rich in yellowish-brown glass, but the grained ones also contain important quantities of it.
The grained rock consist of albite-twinned microphenocrystals of plagioclase corresponding to labradorite, with a feeble zonar struc- ture; the measurements gave an anorthite content lying between the extremes of 43°/o and 70°/o, the greater part indicating narro- wer limits: 56% and 60% An. The needle-formed microlithes of the groundmass generation were determined as corresponding to Ab,,An,,, the limits being an anorthite content of 43 and 55 %o. The contours of the microphenocrystals are sharp whilst the outlines of the microlithes often are rough. These feldspars are dispersed in the brownish glass without visible order or fluctuation arrangement. — A yellowish augite appears also in two generations, the micro- phenocrystals showing distinct prismatic forms corresponding to (100) (010) (110), terminations (001) and (111), and twins after (100), or radiating groups with centrifugal growth, or irregular intergrowths, in some degree influencing the feldspar growth. The optical pro- perties as follows:
City == AD
7—B = 0.0226]
2 V7 = 43°, with axial plane in (010);
axial dispersion around y: A—e =v B-—o>vu
38 ARCTIC BASALT PLATEAU
indicate some enstatite-augitic affinity,! thus FeO and MgO domi- nating over CaO. A feeble zonal structure with decreasing axial angle and birefraction to the centre, and a faint pleochroism in yellowish and greenish tints strengthen this affinity, as dropformed glassy inclusions too, arranged in files parallel to (001). The second- generation-augites are only occasionally developed as minute grains, between crossed nicols on the black ground appearing as lighting points, who increase in importance and dimensions with decreasing glass content. In the greater part of the specimens the second generation of augite is oppressed. — The olivine is apparently crystallized only in the grained rocks, in which the glass is less abundant, and even here it is subordinate, appearing as small rounded grains, mostly pseudomorphosed to a brownish-red, pleo- chroic and fibrous aggregate of ironbearing serpentine (?), optically negative and with relatively high birefraction, and to a reddish-brown isotropic mass, perhaps due to resorption. The grains are rounded and of small size and apparently crystallized after the first appea- rance of augite; in the glassy rock it is quite wanting. — Apatite in small and scarce needles was detected only in rocks poor in glass, where the ore mineral is well individualized as sharp octahedrons of magnetite. In the glassy rock the magnetite forms dendritic ske- leton crystals of minute size contributing to the dark colour of the glass.
Besides the inclusions of sedimentary rocks mentioned above (p. 17) the basalt contains great allogenic fragments of a polysynthe- tical twinned albite-oligoclase (12 °/o An), and of a greenish diopside
of somewhat divergent optical properties (2 Vy—=56#°).
The sphaerules are yellow to dark brown transparent in sections. The smaller ones are isotropic, but in the greater ones they develop radiating structure with birefracting radii, optically negative; this birefraction seems to be in close connection with the progress of
‘9g p. 1—131.
Ay Molecular 3
DESCRIPTION OF ROCKS AND ANALYSES 39
drying of the sphaerule: concentric shrinking cracks appear on them. _ The glassy groundmass of the rock often seems to change into a
substance of similar character, partially when presenting greater masses, also with sensible birefraction and within a greater scale of colours. On the other hand, the glassfchanges apparently in an other substance too, having some community with. the serpentine substance mentioned above; this alteration process is characterized by more greenish tints. — The matter of the sphaerules presents some resemblance to the „hullite“, and is described by various
‚authors under different names; it does accompany basalt and dia-
base flows of several types.’ Around the sphaerule cavities tangen- tially placed feldspars form a complete circle.
A specimen of this rock relatively poor in glass ana apandoned by sphaerules was analysed with the following results:
| 3 A num- er- ; i “ | ae Mer Dt | Mineral composition („Nerm“) | of Osann | SiO, 2:51.60. 49.12 || O.860 57.281 ÖR NN 2.70 1s...579| Al,O, 1438911 l'3-62 |: 01146} CE ÖP ace Ne No 8.34 1 Arne Fe,0, | 4.32! 6.78 | 0.026 ‘uo pol ABs nes. 23.58) F DISSE NES PEO 960 MATS 0131 Ais i Aree ee test 23.91 WISE MgO | 6.09) 3.19} 0.152 | 10.12 | Serene Keio haa Ge | L 58.5: | _ CaO 6.77 | 8.70] 0.120} 7.99 | oH 2 la... 38 | Na,O | 2.78) 2.49} 0.04) 2.99) Di sCaSiO, ... | LER 2:5 K,O | 1.39} 1.26] 0.015 | 1.00 MgSiO, ... 4.06 PRATER H,Oat110° 0.87) fe IMESIO, - al 2... 06 0. y | H,O > 110? | 1.33 J DE — (FeSiO,. . . 12.67 |M.. 2.26 | TiO, 0.00 0.80| 0.009, 0.60 | ME ........... 68 ay 7.es | P,O, aa a CO a Eee 1.375 | ‚ MnO ee na Z fem 39.56 | | | | DO een oer 0.03 | | | | KON: 2.20 100.27, 99.58, 100.00) 100.03 | We | Oo) 6 ie Magmatie symbol: Ul, 5, 3,4 | | | | Camptonose |
1 A chemical analysis of the isolated sphaerules confirms the closeness of them to „hullite* as shown by the comparison tabulated as below: (next page)
40 | ARCTIC BASALT PLATEAU
3. Basalt, Adzva river. H. Backlund, analyst. A. , King Charles land, Spitsbergen. Dr. N. Sahlbom, analyst. !
No volumetrical determination of the mineral constituents was undertaken, because the amount of glass varies from place to place in the same specimen, and no supports of a uniform composition of the glass are offered: the fluctuating content of magnetite skele- tons in the glass residue seems to corroborate this supposition.
A comparison of the two analyses tabulated above (3 and A) shows the close resemblance of them; they belong both to the camptonose subrang, contain nearly equally high quantities of MgO--FeO and are both characterized by a pyroxene of the en- statite-augite-group, this constituent being extremely well developed in the basalt of King Charles land and the olivine mineral being completely resorbed,? whilst in the Adzva rock the more glassy consolidation interrupted a development in the same direction. This resemblance seems to be in no part an accidental one, but seeks.
Gah — |, Sphaerules from the basalt. H. Bac k-
I | II | lund, analyst. | Il. Hullite from the Kinkell-basalt, Scot- | SiO, Sl age land. Heddle, analyst. (Transact. Edinb. Al,O, Toney fee EA Roy. Soc. 1879. 29. p. 89.) MOR 2.41] | Ku | FeO ee MgO 8.00 8.65 CaO fe Drouin). Brod | Na,O er none? | K,O 0.19 | 0.67 | HzO.at 110° | 15.87 8.04 OLS OKO ears MOS hela 1348 | TiO, | 0.18 | not det. | | MnO | not det. | 1.56 99.4 100.
1 A. Hamberg, Die Basalte des König Karls Landes. Geol. Fören. i Stockholm Förh. 1899. 27. p. 509.
2 H. Backlund, Über einige Diabase aus arktischem Gebiet. Tscherm. Min. u.. Petr. Mitt. 1907. 26. p. 378.
ont
i j
MODE OF EXTRUSION 41
her origin probably in a similar geological history of these parts
of the Petehora plain in the sense of Suess (cf. p. 16); and a fur-
ther consequence is a close concordance in time of eruption of both rocks. — Already Holmes fixed the basalts of King Charles and as being a link of the Arctic ones from a chemical standpoin t, and the writer once raised (l. c.) their common character with the diabases of Eastern Spitsbergen in large content of enstatite-augite ; perhaps the age of their extrusion is somewhat later than the last mentioned, thus the low content of titanium dioxide do be constant from King Charles land to the Petchora plain and lower than in the _ older Spitsbergen rocks, lower than in the arctic basalts in general. No alkaline affinity seems to appear in this western part of the
_ eastern plateau-half.
3. MODE OF EXTRUSION OF THE EASTERN BASALTS.
While the greater part of the Atlantic half of the Arctic plateau is buried beneath the sea, in consequence inaccessible to detailed investigations upon the mechanics of extrusion and not suitable to critical comparisons from locality to locality the geological fate of the parts being too variegated in their conditions, contrarily the Siberian half exposes a vaste land-suriace, which in her central part since proterozoic time has undergone no folding processes and the- efore may offer very uniform conditions with respect to magma- extrusions. |
It is true that the part of the vaste plateau lying north off the Angara (Upper Tunguska) river is a difficult task to investigate and the general exploring ways follow the heading stream valleys, thus often giving a wrong idea of the geological structure of the tract between them. The beginning systematic geological mapping on the southern part of the area (Meister) affirms this statement, for
42 ARCTIC BASALT PLATEAU —
the capping basalts in this part have a much lesser spreading than may be concluded from the sections on the Angara river, asides from the principal valleys the basalt failing quite.
The points of outburst of the basalts are unknown. The ancient „volcanoes“ pointed out in the upper Lena river region (by Gue- rassimoff a. 0.) play a subordinate role and have not any impor- tance upon the wider distribution of the basalts being local short- lived outbreak-points of lava without considerable tuff production.
A closer examination! of the basalt-lined valleys of the north reveals at irregular intervals oblique or vertical offsets of basalt dikes of different thickness, which do not in general produce any greater disturbance in the horizontal strata than a local upbending, occasionally witn included greater rock fragments; on both sides of such an upbending the strata redeem their original position at a short distance from the dike without any relative displacement being visible. These fissures seem to have no greater connection with structural lines in general and they get lost in the vaste bordering plateau land without any sign of volcanic action except the related one on both sides of the river bed, at some height from from the bottom and at some distance downstreams. - This statement is valid for broader valleys of greater age. On the tributary valleys of lesser importance and greater youth the basalts fail and the canion-shape disappears. If the offsets of basalt are arranged closer together, the basalt beds corresponding each to an offset join building up a uniform formation of constant level, without visible discordance, and line the canon-shaped valley on both sides. If, in exchange, the offsets are distant each from the other, the tracks of volcanic action are less significant, the belonging basalt bed rests within small dimensions and is oiten eroded away, leaving a basaltic tres-
1 The following statements are in general based upon observations made by the writer during several crossings of the plateau north off the Lower Tunguska river (cf. H. Backlund, „La Géographie" 1907. Paris) and upon studies of travel- ‚reports (Maack, Czekanowski, F. Miiller a. o.). °
MODE OF EXTRUSION 43
hold on the) bottom of the river bed and sometimes small basalt cliffs on one or on both sides of the valley.
The outbreaks are thus fissural ones, tuffs and volcanic agglo- merates are not quite wanting, but they are scarce and saved from erosion only on several points of smaller importance by reasons of ‚conditions related below. It may be supposed strange, that basalt beds of such a mightiness may influence the sedimentary series to such a small degree, but even in center points of basalt develop- ment the bordering palaeozoic strata are by no means, if at all, dislocated in such a degree as would be expected by a single out- break of these dimensions: the strata on both sides of such a mighty „basalt stream“ correspond roughly each to the other stratigraphi- cally and in relative height.
A travel through these countries along the principal rivers — the only ways of communication in summer time — crossing miles after miles central areas of basalt development leaves the impression of a uniform basait plateau, due to one or several concordant lava flows of important mightiness, without any eminent physiographical peculiarities; in retiring from the central areas the river erosion eats through the lava cap and the floor rock appears in undisturbed position; the lava bed thins out, but always crowns the upper edge of the valley and the valley broadens immediately. A cross section through such a valley with uniform basaltic slopes reveals behind the upper edge an ideal plateau of almost absolute horizontality, strewn with lichen-covered heavy rock-débris; the basalt persists only in the immediate vicinity of the upper edge and then sedi- . mentary rocks of the main palaeozoic type replaces it. The boun- dary line between basalt and palaeozoic limestone (marked by a different flora of lichens) runs roughly parallel with the main valley. In places, the basalt edge continues outward to a narrower or broader horizontal plateform, from which a gentle slope leads up to the true plateau. The mentioned boundary runs then at the base of this slope; the contact line, even here covered with rock-débris, is more
44 ARCTIC BASALT PLATEAU
easily visible and its conformity with the valley direction is not difficult to. establish. This matter of fact may awake the thought, © that the basalt streams follow ancient river beds, the rapid broadening of the valley downstream immediately out of the basalt region giving some support to this thought. The parts of the valley cut in the basalt are in some sense antecedent.
li the thought developed above were true, the deepseated broad canon-shaped valleys may have been cut down through the basalt measuring in some parts 300 metres and more of thickness, that being the depth of some valleys and such a mighty outfilling of the valley might seem improbable. The question arises, why did the river course not change passing in the casing, less resistant calcareous sediments? ;
Previous explorers have certainly reported that the basalt series were composed of a number of separate flows, crowded one upon another with astonishing regularity and of various thickness; that it often happens, that the crowning flow is accompanied in lower level by a parallel intrusion sheet (or sill) of constant thickness, the verti- cal distance between the capping basalt-bed and the sheet being almost constant too.' The massiveness of the sedimentary series. may let such a constancy appear astonishing. The old name „trapp“ = staircase) for siberian basalts suggests also a parting in a number of separate flows. In this case it seems probable that the flows may have poured one upon the other, the lowest being the oldest and the topping bed the youngest.
In broader parts of the valleys this partition in mighty flows is well developed and the niveau of each flow persists in horisontality and great constancy, the surface of each of them forming terraces. The shapes of these terraces reminds one from a distance of waterborn raised beaches, and this resemblance leads the thought to updammed lakes with elevated shore lines. The most remarkable fact is that while
1 An apparent geological structure of identical character was tested also by the writer.
rd
MODE OF EXTRUSION | 45
the uppermost lava bed is coarse grained, without sensible marks of surface flow, the lowest stream in exchange reveals an amygda-
loid structure and a very fine grain, even pumice being occasionally encountered. The evolution of columnar structure too, where en- countered, is somewhat paradoxal; on the top of the basalt pile the columnar development is heavy, the individuals being short and of
important diameter, whilst the bottom flow sometimes, when ever
they have such, exposes exellent groupings of slender pillars and feather-shaped radiating arrangements, which point out an indubi- table dependence of the actual surface of the valley bottom, which may consist of basalt or of palaeozoic sediments.
The narrower parts of the valleys, where the detritus cones are swept away, sometimes expose the sedimentary country rock under the top basalt, the floor of the latter being a surface of iluviatil erosion, in some places with waterworn pebbles of sedimentary origin baked together. The sedimentary stripe is by no means the roof of the next basalt horizon below, as may appear from a dis- tance, but the contact plane, apparently almost horizontal, dips teally beneath the basalt flow obliquely down to the valley axis, thus representing the slope of the valley in a newer cycle of ero- sion. This fact, observed in a couple of localities, suggests, that the greater part of intrusive sheets (or sills), related by former ex- plorers, may be only apparent ones,! yet really represent extruded basalt flows, which follow the valley bottom at the end of
a new cycle of erosion. River pebbles, both of sedimentary and
basaltic origin, found somewhere at the base of this second flow and,
where exposed, at the bottom of the following one, strengthen this
thesis. The basaltfilled valleys expose, in their present Stage, an inverted sequence of basalt flows; the oldest flows crown the upper edges of the valleys, while the youngest flows cover the valley floor, apparently forming
1 The massiveness of the sedimentary strata, without well developed bedding panes, is not favourable to intrusions of such regular sheets or sills.
46 SIERRA DE UMANGO S
that the lower permian tillites rest directly over this twice graniti =
basement.
As was already stated in short, the white granite-syenite intrusions - are met with principally in the western parts of the Umango area d outcropping as broader zones in the Sierra de Umango ridge and
“
the Cerro Potrero Viejo. The whole intervening space between the former ridge and the Cerro Cacho is also rich of such dikes. inf the last mentioned mountain there are similar dikes too, but a rede granite, however, dominates. Only in one -part the intrusion of white granite has an abyssal character, viz. in the northern end of the Sierra de Umango ridge. Generally the granite bands are folded together with the limestone banks without crushing, and this eircum- stance makes it evident, that the intrusion was contemporaneous with the tangential pressure. 4
The folding and the banding are also phenomena characterizing the appearance of the white granite-syenite. To the east, in the region of the Famatina batholith there are no signs of lateral stress, and a younger granite seems to be lacking there too. Boden- bender mentions no occurrence of this intrusive from there, and as to myself, I have never discovered any dikes of this rock inside ” the batholith. ;
Nothing further are known regarding the occurrences of youn- | ger (post-Famatina) granites in the pampean sierras-area, be- cause the study of the chronology of the old intrusions in this area
—- we
is still a task never touched upon. According to a verbal com- — munication by Rassmuss, he has seen traces of a younger intru- j sion in the Sierra de Velazco, a white granite crossing the red one, the i latter corresponding to the Famatina granite.
a) MANNER OF INTRUSION OF YOUNGER GRANITES AND SYENITES.
Os fe 8 Me
As already stated, the younger intrusives show a less conspicuous abyssal character than the Famatina red granite, and they depend ~ to a considerable degree in their manner of intrusion on the struc- —
CRYSTALLINE BASEMENT 47 “tural planes of the schist mass. Crosscutting dikes are occassionally “seen, but more irregular bodies with sinuous contacts and numerous _apophyses are not met with. Brecciation, melting and assimilation are also odd phenomena, except a certain class of banding, ‘where a lateral assimilation on a very reduced scale has suc- ceeded. The whole appearance of the intrusion depended appa- rently upon a lateral controlling pressure, causing the mentioned parallel injections along the planes of weakness. The areal exten-
sion of the resulting banded rocks is a very considerable one. Alt- hough they are of the same origin as many common parallel ar- terites in archaean terranes, the banding in the Umango area shows ‘a somewhat different character, due to an already mentioned lateral assimilation along the granite veins. The lateral assimilation is com- mon especially where the granite intrudes amphibolite schists, from which results a product of intermediary composition. This last is also banded, evidently a result of parallel movements under stress conditions. Also at the borders of other schists such a banding was produced.
The banding phenomenon has been studied already long time ago
in different archaean terranes. The following two classes may be distinguished: 1) Banding produced by flowing in a primary magma with the formation of „Schlieren“ or streaks of different composition. The well known banded gabbro of Skye, Scotland, described by Geikie and Teall! belongs to this type. 2) Banding produced by the injection of magma along the structural planes of a schist mass. This type is the more important one and is of regional extension, as is shown by investigations in many parts of the world.?
oo 1A. Geikie and J. J. H. Teall: Banded Igneous Rocks of Skye. Quarterly Journal of the Geological Society of London. Vol. L. 1894. 2J. J. Sederholm: Om granit och gneis. Bull. Comm. Géol. de Finlande. N:o 23. Helsingfors 1907. P. J. Holmquist: Zur Petrographie und Geologie von Ornö Hufvud. Bull. Geol. Inst. of Upsala. Vol. X. 1910. Adams and Barlow: Geol. of the Haliburton and Bancroft Areas, pro-
48 ARCTIC BASALT PLATEAU
quence of the basalt outbreak period and geological incidents to be rela- ted below a capturing of river heads on a large scale took place, partly changing the main drainage in general and completely changing it in details. Thus this ribbon of mesozoic deposits represents an ancient river course (with affluents) of mesozoic age, which drained the central-asiatic mesozoic (Angara)-continent. From the top of the main plateau (e. g. from a point at 68° N and 102° E) there are often seen, generally in the direction of les- sened basaltic action, smaller isolated table mountains, which seem to surpass in height the surroundings. They are crowned by a basalt cap and the slopes are completely covered with basalt-debris, which do not permit any conclusion upon the base of the roofing basalt.! The connection of the basalt with the main flows is un- certain. On the slopes, but never on the top-basalt of these table mountains there were encountered waterworn pebbles of palaeozoic rocks, which in this strange company and ‘at this exposed point
‚lead the thought to glacialdrifted erratics, no other vestiges of gla-
ciation being recorded. A development step by step of the percep- tion reported above forces to analogical conclusions with respect to these table mountains too, that they represent the bottom of ancient river courses, perhaps affluents to the main system, filled out by a basalt flow; the sedimentary rocks served as frame. In sequence the lesser resistant country rock of the frame was strongly denuded and the more resistant basalt flow was brought into relief by its greater resistance to the denudation agents; the pebbles thus derive from underneath the basalt cap. The landscape was morphologi- cally inverted. This statement affirms that far from all water courses, even principal ones at the present time, have persisted as such from mesozoic time, and that in many cases, when the valleys were filled out by basalts, the river courses eroded a new bed sideways of the old basaltfilled valley, no „valley-in-valley“-structure having been developed. |
1 „Witnesses of erosion".
PUST
MODE OF EXTRUSION 49
Much has been written about the contemporaneousness of these basalts. Suess (l. c.) states, that while some flows crown the upper edge of the valleys, others have run down along the bottom; from | this he concludes, that they are not contemporaneous in a stronger sense, but owe their outbreak to a uniform geological process. Of this process he makes no decisive statement, yet he enumerates with coascientiousness the presumptive „volcanoes“, or centres of eruption, and mentions the possibility of fissural one.
On base of the Statements above it is possible to give a picture of the mode of extrusion of the basalts confirming the uniformity of the connected geological process as follows. Towards the end of the Angara-time the continent being almost completely base- levelled began to rise. The river erosion immediately reinforced began to cut down deep valleys, a stagnation of the rise contributed to a development of broad river plains whither the flora migrated down, whilst the capping younger sediments of the uplifted area began to be eroded away.
At that stage the first basalt lava breaks out from fissures, which seem to set up from downward, under an upward pressure, and these fissures meeting on their way through the uniform sediment plateau the deep-seated valleys open and pour out a portion of lava, but seem never to reach the main surface, or if so, the im- mediate output of basalt in the valley seems to hold the main lava surface on a level under the main plateau surface. If the flow was uniform, the river keeps back its original course and the erosion is reinforced anew, because an increasing rise, besides the uplift of the river bed by the basalt layer, immediately succeeds the basalt eruption, continues during the accelerated throughcutting of the basalt and of the floorrock and dies away by forming a new but somewhat smaller river plain. Then the process is repeated with alternating basalt outbreaks and maxima and minima of uplift.
Not everywhere the maximum of upheaval seems to be equal. Yet the maxima of upheaval do by no means coincide with the
30 ARCTIC BASALT PLATEAU
maxima of volcanicity, the latter being in high degree dependent on the-depth and number of the dissecting valleys. Comparing the absolute heights of the plateau, and combining them with the migh- tiness of the basalt piles and with the number of separate flows a
‘ line may be drawn, which runs from the upper Piasina region in
the north to the Tunkin Alps in the south and which may in gene- ral correspond to an axis of the greatest upheaval (fig. 2). This axis of the greatest upheaval coincides to a great extent with the ancient valley depicted above and hence the great development of volcanicity along this line may be explicable.
@ Basalt e— + Axis of maximal upheaval
Fisk
As to the cause of these upheavals the thought of isostatic read- justment lies at hand. Yet there is no distinct region of subsidence in the neighbourhood, that could be taken into consideration as a compensating area, the nearest lying too far away and being separated from the plateau described above by broader zones of tectonical weak- ness. Moreover the uniformity and the flatness of the (northern) area does not favour such a strong and accelerated denudation, as may be : expected in areas of uplift. In all, a rhythmical uplift, at well defined intervals, may hardly find its explication in isostatic readjustment alone.
CONCLUSIONS 51
a De Geer discussed the late tertiary uplift of Scandinavia and Ee pointed out, that in one third the border of the region of upheaval ;
is accompanied by basaltic eruptions, which were pressed up by the downwarping of the adjacent sea-bottom,! by manner of hydrostatic pressure; in sequence the eruption channels were stopped up and the
pressure transplanted to the main backland resulting in an increasing
upheaval. In the case of the eastern half of the Arctic plateau, the magmatic action beneath the plateau seems to be a more
active one than was to be expected by simple isostatic eadjustment, and perhaps reactions and movements in the -asthenosphere, ? as suggested by Ampferer, ? may be the true
cause of the vertical crust-movements in the case considered.
In consequence of the statements above, the basalt-marked line, which crosses the lower Khatanga, Anabar and Olenek rivers and almost points out the boundary between palaeozoic and mesozoic
_ marine sediments, represents a step fault as described by the writer
before," but characterized by upheaval of the southern wing, not
by downwarping of the northern one. The volcanics of this line are
not yet sufficiently known, but seem to be of a more variegated kind.
6. CONCLUSIONS. ©
The eastern half of the Arctic basalt plateau is characterized by volcanics representing the mixed features of the arctic petrographical province; in general, they agree well with the western ones, showing
almost a constant and uniform composition over great areas, and
then accidentally acquiring pronounced signs of an alkaline affinity,
and developing a greater variety of rocks.
1 G. De Geer, Kontinentale Niveauveränderungen im Norden Europas. Pet. Mitt. 1912. IL.
2 A. Holmes, Radioactivity and the earths thermal history. Geol. Mag. 1916, p. 267.
3 ©. Ampferer, Uber das Bewegungsbild in Faltengebirgen. Jahrb. geol. Reichs-
anstalt Wien. 1906. 56, p. 606 et seq.
4H. Backlund, „La Geographic“ 1. c.
d2 ARCTIC BASALT PLATEAU
The alternating eruption took place during a distinct period of rhythmical continental upheaval in such a close alliance the former with the latter, that it is difficult to separate the cause from the
effect. The mode of eruption, being a fissural one, points toward
a deeper magmatic action as the cause of upheaval and eruptivity; the fissures, which generally do not represent faulting lines, were opened from downward by upward pressure, and, meeting on their way upward stripes of pronounced weakness formed by ancient deepseated valleys, they do not reach the main plateau surface, at least not by their volcanicity; the latter finds its outlet in pouring down the valleys. The plateau behaves as a uniform block, and the uniformity of volcanics seems to stay in connection with this beha- viour. In places where the fissures develop into veritable faul- ting lines, with differential and opposed movements on both sides of the line, the volcanics seem to become more variegated, specially in cases where the faults are manyfold and the blocks between them become tilted (New Siberian Islands with Bennet Island).?
It is to be controlled if this statement is to be valid for the At- lantic half. Few significations may satisfy.
De Geer states, in his memoir upon the physiography of Spits- bergen,” that a zone of diabasic eruptions trends „from SE to NW as a diagonal through the region of the present Ice Fjord, separa- ting the main fjord from the diverging four inner branches‘. The diabases are very uniform ones, and this zone marks a distinct flexure of upheaval. The Stor-fjord line of diabases, running in
1 From the standpoint developed above even simple faults, accompanied by differential movements and volcanic action, may, in consequence, depend upon movements and actions of a different kind in the subcrustal magma (astheno- sphere) as compared with the upward movements beneath the plateau; and con- cluding from uniformity of force to homogeneity of acting material the logic compels one to suppose a diversity of deepseated material in connection with irregular mo- vements; thus the alkaline character of rocks seems by no means to be obliged to assimilation of country rocks, but can appear independent.
2 G. De Geer, On the physiographical evolution of Spitsbergen. Geografiska
Annaler 1919. 2. pag. 164 (Stockholm).
Re ee
3 oe CONCLUSIONS CE 53
4 N-S direction along the eastern shore of the fjord, marks also a.
zone of upheaval and the basalts (diabases) are confined to trans-
— versal fissures; the northern straits (Hinlopen and farther north) are
areas of distinct upheaval.
4 Reck, in his large monograph on icelandic mass-eruptions,! tries
4 in vain to prove the „horst“-origin of the steep, vertical-sided tabular
mountains of minute square area (Herdubreid a. o.), the whole res-
ting area around them being broken down by faults. The thesis appears so improbable, that the author himself does not believe it: their origin by upheaval is quite sure. In all, the faulting pro-
_ cesses related from Iceland must undergo a new control from the
| standpoint of discernment between movements of upheaval and sub-
q sidence, for if all subsidences related from Iceland were true ones,
: there would not remain much of it over the sea level.
The consequence of the views developed in course of the present notice is, that volcanics, as once raised by the theory of v. Buch and his followers, may be more active than is admitted even by Gilbert in his laccolith-theory, the degree of moving forces lying somewhere between both ones.
Geological Institution of the Academy of Åbo. May 1920.
1 H. Reck, Isländische Masseneruptionen. Geol. u. palaeontolog. Abhandl. her- ausgegeb. von E. Koken. N. F. IX. 1910. H. 2, pag. 30 a. o.
ACTA ACADEMIAE ABOENSIS MATHEMATICA ET PHYSICA I:3
ÜBER EINE DURCH EIN DREIDIMENSIONA- LES RESONATORENSYSTEM ERZEUGTE INTERFERENZ DER ELEKTROMAGNE- 2 TISCHEN WELLEN
" VON
= DER KAREL Re aN) MAN
PROFESSOR DER: PHYSIK AN DER AKADEMIE ÄBO
ÄBO AKADEMI ÄBO 1921
INHALT:
z ; ra ae |
sits
as pee
FAL ANZ
ee
I. EINLEITUNG.
1. Als ich im Winter 1909—1910 im physikalischen Institut der Universität Helsingfors mit Untersuchungen über das selektive Ab- sorptions- und Reflexionsvermégen elektrischer Resonatorensysteme! beschäftigt war, benutzte ich bei einigen Versuchen dreidimensionale
_ Gitter (Raumgitter), die aus mehreren zueinander parallelen, äqui-
distanten ebenen Gittern bestanden. Als Elemente der Gitter dienten
tingiérmig gebogene Metalldrähte von derselben Eigenschwingungs- periode, welche in horizontalen bezw. vertikalen Reihen angeordnet waren und von dünnen aus Pappe veriertigten Leisten getragen wurden. Beim Messen der Schwächung, welche die mit den Eigen- schwingungen der Gitterelemente angenähert isokronen elektromag- netischen (Hertz’schen) Wellen beim Durchgang durch ein solches Raumgitter erlitten, beobachtete ich gelegentlich, dass die durch- gelassene Strahlungsintensität sich sogar verringern konnte, wenn jedes zweite der aufeinander folgenden ebenen Gitter entfernt wurde, und nach Einschaltung dieser Gitter sich wieder vergrösserte. Es handelte sich hier offenbar um eine von den an den verschiedenen Netzebenen reflektieren Wellen erzeugte Interferenzerscheinung. Nach dem Entfernen der zwischenliegenden Gitter näherten sich im der Tat die Abstände zwischen den an ihren Plätzen gebliebenen ebenen Gittern der halben Wellenlänge der senkrecht einfallenden Schwin-
1 Die Ergebnisse dieser Untersuchungen sind niedergelegt in meiner Abhandlung „Über das selektive Absorptions- und Reflexionsvermögen elektrischer Resonatoren- systeme“ (Öfversigt af Finska Vetenskaps-Societetens Förhandlingar, LII, 1909—1910,
A, N:o 10).
6 INTERFERENZ ELEKTROMAGNETISCHER WELLEN
gungen, so dass die an diesen ebenen Gittern reflektierten Wellen sich dann durch Interferenz verstärkten, welches — trotz der gerin- geren Anzahl von Netzebenen bezw. Gitterelementen — eine ver- grösserte Schwächung der durchgehenden Strahlung zur Folge hatte.
Da der Zweck der erwähnten Untersuchungen ein anderer war, untersuchte ich damals diese Interferenzerscheinung nicht näher. Irgend welche Beobachtungen über Interferenzerscheinungen bei Raumgittern dürften auch kaum früher gemacht worden sein. Zwei Jahre später (1912) folgte indessen die berühmte Laue’sche Ent- deckung der durch die Atomgitter der Kristalle erzeugten Interferenz der Röntgenstrahlen.
Da nun die Hertz’schen Wellen — trotz ihner enorm viel grös- seren Länge — mit Bezug auf ihren impulsartigen Charakter eine gewisse Analogie mit den Röntgenstrahlen zeigen und die Bragg’sche Interprätierung! des Laueeffektes sich auf der Annahme der Existenz äquidistanter ebener Atomschichten in den Kristallen gründet, an denen die Röntgenstrahlen reflektiert werden, scheint es nicht ohne Interesse, die oben erwähnte durch dreidimensionale Resonatoren- systeme erzeugte Interferenz der Hertz’schen Wellen etwas näher zu untersuchen, zumal da ein elektromagnetisches Analogon zu der Röntgenstrahlinterferenz in Kristallen dadurch eventuell erhalten werden könnte. Eine solche Untersuchung hatte ich schon vor vielen Jahren geplant, obwohl ich wegen anderer wissenschaitlicher Arbeiten und verschiedener dazwischen gekommener Hindernisse sie erst jetzt zur Ausführung bringen konnte. ? Die Ergebnisse meiner diesbezüglichen Versuche werden in dem vorliegenden Auf- satze mitgeteilt.
1 W.L. Bragg, Nature 90, 410 (1912); Proc. Cambr. Phil. Soc. 17, 43—57 (1913).
2 Nach meiner Übersiedelung nach Abo (im Jahre 1918) zeigte es sich, dass der Grundboden des Gebäudes, wo das physikalische Institut der hiesigen Akademie eingerichtet wurde, so instabil war, dass sogar die Julius’sche Aufhängevorrichtung eine erschütterungsfreie Aufstellung der von mir benutzten Spiegelgalvanometer nicht ermöglichte. Erst nachdem das Institut vor kurzem ein Annex in einem an- deren Teil der Stadt erhielt, konnten die Messungen dort ausgeführt werden.
EINLEITUNG 7
Da im Folgenden Hinweise auf die Untersuchungen der Herren
Bragg! öfters gemacht werden, scheint es dienlich, die beiden fun-
damentalen Gesetze hier anzuführen, die von diesen Forschern für die Reflexion der Röntgenstrahlen an Kristallen zunächst aufgestellt wurden:
|. Einfallswinkel = Reflexionswinkel.
2. Wenn homogene („monokromatische“) Röntgenstrahlen auf die Kristallflache einfallen, erreicht die reflektierte Intensität nur für
_ singuläre Einfallswinkel (i) merkliche Werte und zwar lassen sich
diese Winkel durch die Formel 2dcosi=nA
darstellen, worin n relativ kleine positive ganze Zahlen (die Ord- nungsnummern der Reilexe), d und A Konstanten sind. Nach der Bragg’schen Theorie bedeutet d den Abstand zwischen den zur
_ Oberfläche parallelen äquidistanten und mit Atomen in gleicher
Weise besetzten Netzebenen im Kristalle, während A sich als die Wellenlänge der Röntgenstrahlen herausgestellt hat. Anstatt des Eintallswinkels (i) bedienen sich die Herren Bragg gewöhnlich des Neigungswinkels („glancing angle“) zwischen den einfallenden Strahlen und der Kristallflache (bezw. den Netzebenen im Kristalle). Nennt man diesen Winkel q (= 90°—i), so geht die obige Formel über in | 2dsing=nA.
I. APPARATE.
2. Erreger, Empfänger (geradliniger Sekundärleiter mit Thermo- element) und Messungsmethode (gleichzeitiges Ablesen eines Broca’- schen und eines du Bois-Ruben’schen Spiegelgalvanometers) waren dieselben, welche ich bei meinen in Helsingfors über elektrische
1 Eine zusammenfassende Darstellung dieser Untersuchungen findet man in der
Arbeit: W. H. und W. L. Bragg, X-rays and Crystal Structure. (London 1918).
8 : INTERFERENZ ELEKTROMAGNETISCHER WELLEN
Wellen ausgeführten Versuchen benutzt hatte.! Das Arbeitszimmer hat relativ grosse Dimensionen (Länge etwa 9 m, Breite 5 m, Höhe 3,2 m), so dass Störungen durch Reflexionen an den Wänden und dem Fussboden leicht vermieden werden konnten.
Die Resonatoren, welche die Elemente der zu untersuchenden Gitter bildeten, hatten die Form von nahezu geschlossenen Ringen und bestanden aus 13 cm langen und 0, mm dicken kupiernen Drähten. Der Abstand zwischen den Drahtenden betrug 3 a 4 mm. Wie ich in meiner in der Einleitung (S. 5) erwähnten Arbeit fest- gestellt habe, hat ein ringförmiger, nahezu geschlossener Sekundär- leiter von konstantem Querschnitt eine ein wenig grössere Eigen- schwingungsperiode als ein geradliniger von derselben Länge und Dicke und zwar nimmt dieser Unterschied zu, wenn man den Ab- stand zwischen den Enden des ersteren verkleinert. Es erwies sich indessen, dass die halbe Eigenwellenlänge der erwähnten 13,0 cm langen und 0,7 mm dicken ringformigen Resonatoren, deren freie Enden 3 a 4 mm von einander entfernt waren, nur etwa 3 mm länger als die Drahtlänge war (also = etwa 13,3 cm).
Diese ringförmigen Resonatoren waren stets in äquidistanten hori-
zöntalen bezw. vertikalen Reihen angeordnet und wurden dabei von
aus dünnen Holzleisten gemachten Stativen getragen (fig. 1). Die wagerechten 0,; cm dicken Leisten (von quadratischem Querschnitt) waren mit kleinen hölzernen Haken versehen, an denen die ring- förmigen Resonatoren so aufgehängt waren, dass die Öffnung zwi- schen den Enden jedes Resonatordrahtes in derselben horizontalen Ebene wie der Mittelpunkt des Resonatorkreises lag, wobei die Öffnungen aller Resonatoren nach derselben Seite hin gerichtet waren (Siehe Fig. 1). Es wirkte dann bei vertikaler Stellung des stabiörmigen Erregers und bei normaler Inzidenz die elektrische
1 Siehe K. F. Lindman Ofversigt af Finska Vet. Soc. Förh. LI, 1908—1909, A. N:o 5. — Ann. d. Phys. 38, S. 526 (1912). — Ein Teil dieser Apparate (z. B. das du Bois-Ruben’sche Panzergalvanometer) gehért dem physikalischen Institut der Universitat Helsingfors und bin ich Hrn Prof. Hj. Tallqvist fiir die Erlaubnis, diese Apparate noch in Abo zu benutzen, zu grossem Dank verbunden.
2
er x
& 1
APPARATE 9
Kraft der Hertz’schen Wellen auf die Resonatoren. Der Abstand
S zwischen zwei benachbarten horizontalen bezw. vertikalen Reihen der Aufhängungspunkte (bezw. Mittelpunkte) der Resonatoren betrug
10 cm. Jedes dieser ebenen Resonatorengitter umfasste 7 horizontale und 11 vertikale Reihen und bestand somit aus 77 ringförmigen in
derselben Ebene angeordneten Elementen. Das zu untersuchende
Raumgitter wurde aus mehreren (gewöhnlich fünf) solchen ebenen
_ hintereinander angeordneten Gittern zusammengestellt. Dieses Raum-
-epipedférmigen Teile befan-
gitter erhielt dabei die Form eines regelmässigen rechtwinkligen Gitters, in dem die einzel-
nen Resonatoren sich in den Ecken der parallel-
den, aus welchen das ganze GOCCOCTCOC CU Gitter zusammengesetzt gee (OC CO OC CO oe dacht werden konnie. OE Ae
Wenn ein Element eines THON OO Can Gin OM Om Ome) einzelnen ebenen Gitters af
| |
5 5 te : Fig. 1 durch einen ringförmigen 5
Messresonator von derselben Periode ersetzt wurde, zeigte es sich, dass die Wirkung auf diesen Messresonator in Gegenwart der an- deren Elemente des Gitters etwas grösser war als in Abwesenheit derselben. Bei der erwähnten Verteilungsdichte der Elemente hatten aber die anderen Elemente keinen merkbaren Einfluss auf die Eigen- periode des als Gitterelement benutzten ringförmigen Messresonators.
~Beim Messen der Wellenlänge nach der Boltzmann’schen Methode
mit zwei gegeneinander verschiebbaren ebenen Metallspiegeln ergab sich nämlich für die Wellenlänge derselbe Wert, wenn der Messre- sonator ein Element des Gitters war, als wenn er isoliert stand. Aus der Gestalt der dabei erhaltenen Interferenzkurven konnte aus- serdem geschlossen werden, dass die Dämpfung der Eigenschwingun- gen der Gitterelemente bei der betreffenden Verteilungsdichte nur unwesentlich grösser war als die der freien Eigenschwingungen die-
ser (als isoliert gedachten) Elemente.
10 INTERFERENZ ELEKTROMAGNETISCHER WELLEN
Ill. DURCHLASSIGKEITSVERSUCHE.
3. Um die Intensität der durch ein Raumgitter gegangenen elek- trischen Wellen ihrem absoluten Betrage nach richtig messen zu können, kommt es sehr auf eine richtige Abschirmung der ringsum fallenden Strahlung an, damit die Messungen nicht durch Beu- gungserscheinungen vereitelt werden sollen. Da die entsprechenden Reflexionsversuche in dieser Beziehung einfacher und zuverlässiger erschienen und sich direkt mit den oben erwähnten Bragg’schen Versuchen vergleichen liessen, beschränkte ich mich jetzt auf einige wenige Durchlässigkeitsversuche, die eigentlich nur eine Wiederho- lung und Ergänzung meiner früheren in der Einleitung erwähnten diesbezüglichen Beobachtung bezweckten.
Die Versuchsanordnung wird in Fig. 2 schematisch angegeben. Der Oszillator O befand sich in der Brennlinie eines zylindrisch- parabolischen Reflektors, dessen Brennweite gleich !/ı der Wellen- länge der primären Schwingungen war. An dem Reflektor war ein metallischer Trichter 7 von quadratischem Querschnitt befestigt, der vorn eine 33 cm X 33 cm grosse quadratische Öffnung hatte. Der geradlinige mit Thermoelement versehene Resonator /, der sich innerhalb des Trichters in der Nähe seiner Wand und der Öffnung befand, diente als Standardindikator und war deshalb mit dem einen Galvanometer verbunden. Die 5 vertikalen und zueinander paralle- "len ebenen Gitter G,, G,, -- G,, die zusammen das zu untersuchende Raumgitter bildeten, waren auf einer wagerechten 80 cm breiten Holzscheibe derart montiert, dass der Abstand d zwischen den be- nachbarten äquidistanten Netzebenen ohne Schwierigkeit variiert werden konnte. Das dem Oszillator O am nächsten stehende Gitter G, blieb die ganze Zeit in einer unverändertern Entfernung von O. Wenn ein grosser Metallspiegel unmittelbar vor dem Gitter G, auf- gestellt wurde, veränderte sich die Wirkung auf den Standardin- dikator / kaum merkbar. Eine Beeinflussung des Standardindikators
| j :
DURCHLÄSSIGKEITSVERSUCHE 11
durch die am Resonatorensystem reflektierte Strahlung war also nicht zu befürchten. Der mit zylindrisch-parabolischem Reflektor _ versehene geradlinige Messresonator R stand in unveränderter Lage
_ hinter dem Gittersystem. Eigentlich hätte noch eine Blende zwischen
dem Gitter und dem Empfänger angeordnet werden sollen. Für
die wenigen vergleichenden Durchlässigkeitsversuche, die ich jetzt _ vornahm, schien aber eine solche Blende entbehrlich, zumal sie leicht als eine neue Fehlerquelle hätte wirken können. Die mittel-
sten wagerechten Elementenreihen der Gitter und die Mittelpunkte
| I (pe er Alkan i G0 He iy Ban | | | | fx) dididld! 2 \ = J
von O und R befanden sich etwa 1,3 m. über dem Fussboden des _ Arbeitszimmers. Die halbe Eigenwellenlange des Oszillators (1/2 40)
und die des Messresonators (1/2 Ar) waren bei diesen Versuchen beide = 13, cm und also von fast derselben Grösse wie die der Gitterelemente (13,3 cm).
4. Durch Messung der durchgelassenen Intensität beim Variieren
des Netzebenenabstandes d erhielt ich die in Fig. 3 dargestellte
Intensitatskurve. Jede Messung wurde mehrere Male (mindestens
drei aber oft sechs oder sieben mal) wiederholt und aus den so
erhaltenen — stehts nahe miteinander übereinstimmenden — Werten das Mittel genommen. Die maximale Ordinate dieser Kurve liegt bei d = 6,9 cm. Die an zwei benachbarten Netzebenen reflektier-
ten Wellen hatten in diesem Falle einen Gangunterschied von
2.6,» = 13,8 cm, der mit der halben Wellenlänge (13 cm) recht
12 INTERFERENZ ELEKTROMAGNETISCHER WELLEN
nahe übereinstimmt. Das Maximum der durchgelassenen Intensität erklärt sich demnach durch ein entsprechendes Minimum der reflek- tierten Intensität. Ebenso findet man, dass das Minimum der durch- gelassenen Strahlung (für d = 13 cm) einem Maximum der reflek- tierten Intensität entspricht. Wenn ein dreidimensionales aus gleichen ringförmigen in parallelen äquidistanten Ebenen angeordneten Ele- menten bestehendes Resonatorensystem von senkrecht einfallenden, "mit den Eigenschwingungen der Gitterelemente angenähert isokro- nen elektromagnetischen Wellen durchstrahlt wird, so ist also die
120
Durchgelassene Intensität
u S
Netzebenenabstand (d) Fig. 3
durchgelassene Intensität am grössten, wenn der Abstand zwischen den benachbarten Netzebenen gleich "Ja der Wellenlänge ist, und am geringsten, wenn dieser Abstand gleich der halben Wellen- länge ist.
Wenn beim Netzebenenabstande d = 6, cm das zweite und das vierte Gitter (G, und G, in Fig. 2) entfernt wurden, so dass d jetzt = 13 cm (= der halben Wellenlänge) wurde, nahm die durch- gelassene Intensität — trotz der Verminderung der Netzebenen- anzahl — um etwa 6 Proz. ab. Es war dies eine Bestätigung der von mir schon vor vielen Jahren gemachten Beobachtung, die in der Einleitung (S. 5 u. 6) diskutiert wurde.
SATSEN
REFLEXIONSVERSUCHE
IV. REFLEXIONSVERSUCHE.
5. Das vorher benutze aus fünf Netzebenen bestehende Resona- -torensystem, dessen Reflexionsvermögen jetzt untersucht werden sollte, wurde vor dem Oszillator O (Fig. 4) in etwa 2 m Entier- nung von ihm aufgestellt. Der Messresonator R, welcher jetzt mit
N ie \ QE \ A I \ ! Lay Ml / I eligi Led, Lent ett) \ lena Wei erat yl 5 x W _— = nn nam ae m, SE 5 ee ee oe Fig. 4
dem früher (vor dem Erreger) benutzten Trichter 7 versehen war, empfing die am Resonatorensystem reflektierten Strahlen. Der
Standardindikator / befand sich in der Brennlinie eines hohen zylindrisch-parabolischen Spiegels. Ausser diesem Spiegel dienten die Metallschirme S, och S, zum Abblenden der seitlichen Strahlen, wie in der Figur angedeutet ist. Die halbe Eigenwellenlänge des
. Oszillators ("/2 45) und die des auf ihm abgestimmten Messresona-
14 INTERFERENZ ELEKTROMAGNETISCHER WELLEN
tors (!/2 4) war bei allen Reflexionsversuchen, wie auch bei den ~
obigen Durchlässigkeitsversuchen, = 13,0 cm, welcher Wert also um z:a 0,3 cm kleiner als die halbe Eigenwellenlänge der Gitter- elemente war. Diese Differenz, die ich anfangs als unwesentlich betrachtete, war dadurch bedingt, dass nach meinen früheren Ver- suchen! das Gitter diejenigen Wellen bezw. Schwingungskompo- nenten am stärksten reflektierte, deren Periode ein wenig kleiner als die der Gitterelemente war. In Abwesenkeit des Gitters konnte bei verschiedenen in Frage kommenden Stellungen des Empfängers keine oder fast keine Wirkung auf ihn beobachtet werden (es waren also keine oder fast keine merkbaren Fremdwirkungen vorhanden). Beim Drahen des Gitters um eine durch die Mitte des vordersten Gitters (G,) gehende vertikale Axe sowohl in der einen wie in den anderen Richtung ziegte es sich, dass die Wirkung auf den Emp- fänger am grössten war, wenn das primäre Strahlenbündel und das reflektierte, auf den Empfänger fallende mit den Gitterebenen gleich grosse Winkel bildeten. Es wurde dies bei verschiedenen Einfalls- winkeln und verschiedenen Netzebenenabständen beobachtet. Für die Reflexion der elektromagnetischen Wellen an dem Resonatoren- system gilt also das gewöhnliche Reflexionsgesetz, nach dem der Reflexionswinkel gleich dem Einfallswinkel ist. Es entspricht dies dem ersten in der Einleitung (S. 7) erwähnten Bragg’schen Gesetze für die Reflexion der Röntgenstrahlen an Kristallen.
Durch Drehung des Empfängers um seine zu den reflektierten Strahlen parallele Symmetrieaxe wurde festgestellt, dass die reflek- tierten Strahlen, gleich wie die einfallenden, geradlinig polarisiert
waren und dass die zur Einfallsebene senkrechte Richtung des elek-
trischen Vektors der einfallen Schwingungen sich durch die Reflexion am Gitter nicht änderte.
6. Bei den unten zunächst beschriebenen Versuchen war der Ein-
falls- bezw. Reflexionswinkel = 12°. Durch Messung der an dem
1 Vgl. die auf S. 5 zitierte Arbeit des Verfassers.
a , | g 4 | | ;
2
REFLEXIONSVERSUCHE 15
© _ Resonatorensystem reflektierten Strahlungsintensität! beim Variieren
des Netzebenenabstandes d erhielt ich die in Fig. 5 dargestellte Kurve I, die ein Maximum und ein Minimum der reilektierten In- _ tensität deutlich erkennen lässt. Das Maximum entspricht einer
3 Phasendifferenz zwischen den an den benachbarten Netzebenen re- flektierten Strahlen von 2-13,5 cos 12° = 26,4 cm, welcher Wert mit
Se Hd | tA LAS ELE NS RA So ae > ene
POS SI den baer eh RO EZ. Dg a Netzebenenabstand (d)
ae Fig. 5
> Reflektierte Intensität
ee. ware |
der Wellenlänge der einfallenden Schwingungen (26,0 cm) sehr nahe übereinstimmt. Das Minimum der Kurve Fin Fig. 5 liegt bei d=7 cm. Für die entsprechende Phasendifferenz ergibt sich der Wert 2.7cos 12° = 13,7 cm, der um 0,7 cm grösser als die halbe Wellen- länge ist. Die Ursache zu dieser geringen Abweichung, von der wir hier absehen, soll später diskutiert werden. Man findet, dass das _ Maximum und das Minimum der betreffenden Reflexionskurve mit
; 1 Jeder fixe Punkt der Intensitätskurve wurde wieder wie bei allen in diesem
- Aufsatze beschriebenen Versuchen durch mindenstens drei, aber gewöhnlich vier oder fünf (manchmal noch mehr) Ablesungen der beiden gleichzeitigen Ausschläge _ der Galvanometer bestimmt. Die Quotienten dieser Ausschläge stimmten in jedem Falle mit einander sehr nahe überein.
16 INTERFERENZ ELEKTROMAGNETISCHER WELLEN
Bezug auf ihre Lagen dem Minimum bezw. Maximum der Durch-
lässigkeitskurve sehr genau entsprechen, so dass die aus ihnen er- -
mittelten Werte der Phasendifferenz der an benachbarten Netzebe-
nen reflektierten Wellen fast identisch sind. Die vorhin (S. 12) gege- |
bene Deutung der singulären Punkte der Durchlässigkeitskurve wird also durch die Gestalt der Reflexionskurve I (fig. 5) bestätigt. Dass es sich hier in der Tat um eine Interferenzwirkung zwischen den an den einzelnen Gitterebenen reflektierten Wellen handelte, geht auch daraus hervor, dass das vorderste Gitter G, einsam (nach
Entfernung der übrigen Gitter) die Wellen stärker reflektierte (die
gestrichelte Kurve II in Fig. 5) als das ganze Gittersystem, wenn der Netzebenenstand in diesem einen Wert zwischen etwa 5 und 10 cm hatte.
7. Wenn die halbe Wellenlänge des Erregers und die des Empfängers = 15,0 cm waren (Kurve III in Fig. 5), entsprach dem Reilexionsmaximum ein Netzebenenabstand d= 14,7 cm. Auf senk- rechte Inzidenz reduziert, beträgt dies 14,7 cos 12° = 14,4 cm, welcher Wert um 0,6 cm kleiner als !/2 A, (oder !/2 A,) ist. Für ei, = aA, = 15,9 cm (Kurve IV in Fig. 5) ergab sich für den dem Reflexionsmaximum entsprechenden auf senkrechte Inzidenz reduzierten Netzebenenabstand der Wert 15,0 cos 12° = 14,7 cm, der
um 1,2 cm kleiner als die halbe Wellenlänge ist. Für Wellenlän-
gen, welche grösser als die Eigenwellenlänge der Gitterelemente waren, war also der dem Reflexionsmaximum entsprechende auf normale Inzidenz reduzierte Netzebenenabstand des Raumgitters etwas kleiner als die halbe Wellenlänge der einfallenden Wellen. Für !/2 4, =1/2 4, =10,s cm (Kurve V in Fig. 5) entsprach dem Reflexionsmaximum ein auf normale Inzidenz reduzierter Netzebe- nenabstand = 13,7 cos 12° = 13,4 cm, welcher Wert mit der halben Eigenwellenlänge der Gitterelemente (z:a 13,3 cm) fast genau über-
einstimmt. Für eine Wellenlänge, die kleiner als die Eigenwellen-
länge der, Gitterelemente war, richtete sich demnach der dem Re- flexionsmaximum entsprechende Netzbenenabstand vollständig nach
OS en a CE Te NN
REFLEXIONSVERSUCHE : 17
-
der Eigenperiode der Gitterelemente, und bei grösseren Wellenlän- gen war auch, wie wit eben gesehen haben, ein deutlicher Einfluss dieser Eigenperiode merkbar.
8. Die Eigenschwingungen eines stabförmigen eters sind
bekanntlich stärker gedämpft wie die eines ringförmigen von - derselben periode und zwar nimmt die Dämpfung bei Verkür- zung (bezw. Verkleinerung der Eigenperiode) des stabförmigen Leiters schnell zu. Es scheint demnach wahrscheinlich, dass die
Ursache zu den beobachteten Verschiebungen der den Reflexions-
maxima entsprechenden Werte von d darin zu suchen ist, dass die _ Eigenschwingungen des geradlinigen Messresonators stärker ge- dämpft waren als die der ringférmigen Gitterelemente (die Dämp-
fung des mit Funkenstrecke versehenen Primärleiters O war natür- lich noch viel grösser). Bei der Reflexion an dem Resonatorensys-
: tem machte sich also die Erscheinung der „multiplen Resonanz“
_ besonders bei den kleineren Wellenlängen geltend. Unter den - unendlich vielen Schwingungskomponenten, aus denen die primä-
ren Schwingungen zusammengesetzt gedacht werden können, wählte
das Gitter diejenigen aus, deren Periode mit der seinigen überein- stimmte, und reflektierte — wenn die Hauptschwingungsperiode des Erregers kleiner als die Periode des Gitters war — hauptsächlich diese.
Wie ich früher nachgeweisen habe, besitzt ein aus ringförmigen Elementen bestehendes ebenes Gitter ein scharfes selektives Re-
- flexionsvermögen. Die Gitterelemente reagieren dabei hauptsächlich auf die mit ihren Eigenschwingungen isokronen Schwingungskom-
ponenten der primären Wellen und emittieren Sekundärwellen von ihrer eigenen Periode, die sich zu einer Eigenstrahlung des Gitters zusammensetzen.
Es geht dies besonders deutlich aus einigen Versuchen hervor, durch welche die an einem solchen Gitter reflektierte Strahlung durch Resonanzversuche in einem solchen Falle analysiert wurde,
wo die einfallenden Wellen eine grössere Periode als die Gitterele-
18 INTERFERENZ ELEKTROMAGNETISCHER WELLEN -
mente hatten!. Es zeigte sich, dass die reflektierte Strahlung sich dann aus zwei Teilen zusammensetze, nämlich aus einer Eigenstrah- lung (Floureszenzstrahlung), die mit den Eigenschwingungen der Gitterelemente isokron war, und einer zerstreuten oder erzwunge- nen Strahlung, deren Periode mit der der primären Wellen über- einstimmte. Während bei dem aus ringförmigen Elementen beste- henden Gitter die Eigenstrahlung die Hauptrolle spielte, war bei einem aus geradlinigen Resonatoren zusammengesetzten Gitter und . in noch höheren Grade bei einem Kugelgitter die zerstreute (er- zwungene) Strahlung dominierend. Es bildete dieses Ergebnis ein elektromagnetisches Analogon zu der von Barkla und Sadler? nachgewiesenen Zusammensetzung der sekundären Röntgenstrah- lung aus zwei entsprechenden Komponenten.
Wenn bei den oben beschriebenen Versuchen über das Reflexions- vermögen des aus ringförmigen Elementen bestehenden Raumgitters der mit cos 12° multiplizierte Netzebenenabstand angenähert gleich der halben Eigenwellenlänge der Gitterelemente ist, verstärken die von den verschiedenen Netzebenen ausgehenden auf den Messreso- nator wirkenden Eigenwellen einander durch Interferenz und erzeu- gen dann, wie durch die Beobachtung bestätigt wird, ein Maximum der Eigenstrahlungsintensität. Ist die Periode der primären Wellen kleiner als die Eigenperiode der Gitterelemente, so kommt die zer- streute Strahlung neben der Eigenstrahlung kaum in Betracht. Bei grösseren Perioden der Primärstrahlung ist aber die Intensität der zuletzt erwähnten Strahlung von derselben Grössenordnung wie die Eigenstrahlung. Die reflektierte Intensität müsste dann eigentlich bei zwei verschiedenen Netzebenenabständen ein Maximum erreichen. Wenn aber — wie bei den obigen Versuchen der Fall war — der Unterschied zwischen der Periode der Primärwellen und der der
1 K. F. Lindman, Über elektrische und optische Resonanz. Öfversigt af Finska Vet. Soc. Förh. L. VII 1914—1915, A. 2, p. 48—49. Ann. d. Phys. 45, p.620—621 bezw. p. 612—613 (1914).
2C.G. Barkla und A. Sadler, Phil. Mag. 16, p. 550 (1908) — A. Sadler, Phil. Mag. /8, p. 107 (1909).
3 as REFLEXIONSVERSUCHE 19
a u “4
oF N A
- Ce
A
‘Gitterelemente relativ klein ist, vereinigen sich diese beiden Maxima cf zu einem einzigen, wobei der entsprechende (auf normale Inzidenz _ reduzierte) Netzebenenabstand einen Wert zwischen der halben Wellen-
- länge der Eigenstrahlung und der der zerstreuten Strahlung annimmt.
Dass bei den Versuchen mit den 26 cm langen Wellen (1/2 is = !/2 dp = 13,0 cm) sowohl das Maximum wie auch das Minimum
der reflektierten Intensität (Kurve I in Fig. 5) um einige mm nach
_ der Seite der grösseren Netzebenenabstände hin verschoben war,
kann jetzt dadurch erklärt werden, dass die Eigenwellenlänge der ; Gitterelemente (A = etwa 26, cm) ein wenig grösser als die er- - wähnte Wellenlänge der einfallenden Schwingungen war. Dass das
4
Minimum sogar um etwa 2 mm mehr nach rechts verschoben war,
als nach dieser Erklärung zu erwarten ist, ist wahrscheinlich auf
_ Versuchsiehler bezw. eine Ungenauigkeit in der Bestimmung der
Lage dieses Minimums zurückzuführen.
9. Da für Werte von 4, und 4,, die kleiner als die Eigenwel-
_ lenlänge der Gitterelemente waren, die am Gitter reflektierte Strah-
lung nach der oben dargestellten Auffassung hauptsächlich aus der
- Eigenstrahlung des Gitters bestehen musste und diese Eigenstrah-
lung weniger gedampit war als die Eigenschwingungen des gerad-
_linigen Messresonators, bot das Gitter ein Mittel dar, stehende
elektrische Wellen im Luitraume nachzuweisen, deren Länge von
der Eigenperiode des Messresonators unabhängig war. Die dies- bezüglichen Versuche, die gleichzeitig als Prüfung der erwähnten
_ Auffassung dienten, wurden auf folgende Weise ausgeführt.
Primäre Wellen, deren halbe Lange (!/2 4,) gleich 10,8 cm war, fie-
len auf eines der vorhin benutzten ebenen Gitter (halbe Eigenwel- _lenlange der ringförmigen Gitterelemente = 13,3 cm) unter einem - Einfallswinkel von 40°. Die an diesem Gitter unter einem gleich gros-
sen Winkel reflektierten Wellen fielen senkrecht auf einen 80 cm
hohen und 50 cm breiten ebenen Kupferspiegel. Durch Intensitäts- _ messungen mit dem jetzt ohne Trichter und parabolischen Strahlen- _ sammler benutzten auf den Erreger abgestimmten geradlinigen Mess-
20 INTERFERENZ ELEKTROMAGNETISCHER WELLEN
resonator WER = 10,s cm) in verschiedenen Entfernungen von dem 3
Kupferspiegel erhielt ich die Kurve I in Figur 6.
Ein Minimum der elektrischen Kraft lag demnach jetzt in der Ent-
fernung 13,3 cm vom Spiegel, welcher Abstand mit der halben Eigenwellenlange der Gitterelemente genau übereinstimmte. Irgend eines dem Werte !/2 2, = !/2 4, = 10,s cm entsprechendes Minimum ist nicht vorhanden. Die Länge der stehenden Wellen richtete sich somit nur nach der Periode der Gitterelemente, d. h. die vom Gitter
90
Intensität
TERN g lo. VET 2 ES eA Gar
Abstand zwischen Spiegel und Messresonator Fig. 6
emittierten sekunddren elektromagnetischen Welien waren durch die- sen Versuch objektiv nachgewiesen.
Nachdem das Gitter durch einen ebenen Eisenspiegel ersetzt wor- den war, erhielt ich durch einen entsprechenden Versuch die Kurve II in Fig. 6. Die Abszisse des Minimipunktes dieser Kurve ist na- türlich gleich "/24,. Dass das Minimum der Kurve 1 weniger scharf ausgebildet ist als das der Kurve II!, beruht offenbar dar-
1 Wenn man die Verschiedenheit der Höhenskalen der beiden Kurven in Betracht zieht, so findet man, dass der Unterschied in der Schärfe dieser Minima jedoch
nicht so gross ist, wie man nach dem ersten Eindruck von der Figur schliessen könnte,
a re? er ee
=:
‘
= 4 a
REFLEXIONSVERSUCHE 21
aul, dass die Eigenperiode des Messresonators bei der Aufnahme jener Kurve mit der Periode der Eigenstrahlung des Gitters nicht
_ übereinstimmte. Durch Abstimmen des Erregers zur Resonanz mit N den Gitterelementen (}/.A,— 13,3 cm) hätte man eine bedeutend stärkere und wahrscheinlich auch noch weniger gedämpite Sekundär-
N
strahlung erzielen können. Da ich aber die oben für den Fall
Fig. 7
1g 1, = 1/24, — 10,s cm gegebene Erklärung der Wirkung des Raum-
gitters gleichzeitig priifen wollte, sah ich hier von einer solchen
_ Resonanzabstimmung ab.
Durch Benutzung eines Messresonators, dessen Eigenschwingun- gen stärker gedampit waren als die des primären Erregers, habe ich früher! stehende elektromagnetische Primärwellen im Luftraume objek- tiv nachgewiesen.
1K. F. Lindman, Öfvers. af Finska Vet. Soc:s Förh. 1]1908 13097 AL to
Ann. d. Phys. 38 p. 523 (1912).
22 INTERFERENZ ELEK TROM AGNETISCHER WELLEN
10. Ich machte schlieslich noch eine Reihe von Versuchen über ; das Reflexionsvermögen des Raumgitters bei verschiedenen Einfalls- bezw. Reflexionsvinkeln. Um den Einfallswinkel bequem variieren q zu können, war bei diesen Versuchen das Resonatorensystem auf - einem 2 m langen und 0, m breiten um ihre Mitte drehbaren ° Holzscheibe A B (Fig. 7) montiert (wie in der Figur angedeutet ist, ° lag die Scheibe A B auf einer kleineren Scheibe; ein Zapfen M diente ° als Drehungsachse). Vier der vorher benutzen ebenen Gittern waren 4 dabei zu zwei miteinander parallellen um 28 cm voneinander ent- - fernten ebenen Gittern G, und G, zusammengestellt. Um mehrere - Reflexionsmaxima zu erhalten, musste der Abstand zwischen G, und :
G, grösser als die Wellenlänge (26,6 cm) sein, und da andererseits die verhältnismässig geringe Weite (33cm) der Öffnung des Emp-
fangertrichters T nur die Benutzung eines ziemlich eng begrenzten - Strahlenbiindels zuliess, hatte es keinen Zweck, eine grösse Anzahl -
von Netzebenen zu verwenden. Die ganze Versuchsanordnung geht übrigens aus Fig. 7 hervor (die Buchstaben O, R, /, S, und S, ha- ben hier dieselbe Bedeutung wie in Fig. 4). Die Drehungsachse M der
Scheibe AB war mittels dünner Bindfäden mit dem Oszillator O und dem ~ Messresonator R verbunden. Der Winkel &, den die einfallenden und die
reflektierten Strahlen mit den Gitterebenen bildeten, wurde mit Hilie ei- ner Gradscheibe bestimmt. Der Abstand OM war = l,s m und der Abstand RM = 2,2 m. Beim Entfernen der Gitter G, und G, zeigte es sich, dass auch bei dem kleinsten angewandten Werte von & (19°) der Erreger O keine merkbare direkte Wirkung auf den Messreso-
nators R ausübte. Die halbe Eigenwellenlänge des Erregers und die —
des Messresonators waren bei diesen Versuchen = 13 cm.
Durch Messung der reflektierten Strahlungsintensitat beim Varie- ren des Winkels g erheilt ich die in Fig. 8 wiedergegebene Kurve. Obwohl die einzeinen fixen Punkte, zwischen denen die Kurve ge-
zogen worden ist, nicht ganz regelmässig liegen — was vielleicht
auf Störungen durch fremde Reflexionen (der um die Schirmöffnung
gebeugten und am Gitter reflektierten Strahlen) zurückgeführt wer- .
REFLEXIONSVERSUCHE 23
_ den kann — zeigt diese Kurve zwei gut ausgebildete Reflexions- maxima für g = etwa 27° bezw. 68° und ein Minimum für » = 47°.
4 >
Wenn man die zu erwartende Lage der Reflexionsmaxima nach
der Formel 2 d sin g = nA (siehe Einleitung p. 7) berechnet und dabei
4 gleich 26,6 cm (die Eigenwellenlänge des Gitters) setzt, so erhält
man = 28,4 fürn=]1 und = 171, für m= 2, welche Werte in der Tat mit den beobachteten Werten 27° und 68° sehr nahe über-
„ee 90
ee EEE a yt
J0
Intensitat
Neigungswinkel (¥) Fig. 8
- einstimmen. Aus der Formel 2d sin = (n + 1) - ergibt sich für
die Lage des Minimums, wenn man n= 1 setzt, «' = 45,3”, welcher Wert auch nur sehr wenig von dem beobachteten (47°) abweicht. Ohne sie auf Versuchsfehler bezw. eine mangelhafte Versuchsanord- nung zurückzufüren zu brauchen, können die Unterschiede zwischen
den beobachteten und den berechneten singulären Werten von @
(und @’) durch die durch besondere Versuche (siehe unten) festge- stellte, mit wachsenden Einfallswinkeln (90° — ¢) folgende rasche Zunahme des Reflexionsvermögens der beiden Gitter G, und G,
“erklärt werden. Diese Zunahme des Reflexionsvermögens lässt
24 INTERFERENZ ELEKTROMAGNETISCHER WELLEN
in der Tat eine Verschiebung der Maxima nach der Seite der grös- — seren Einfallswinkeln (kleineren Werten vong) hin und eine nach der entgegengesetzten Richtung erfolgende Verschiebung des Mini- -
mums erwarten. Wenn von den betreffenden geringfügigen Verschie-
bungen abgesehen wird, entsprechen also die obigen Beobachtungen | über die Lage der Reflexionsmaxima vollständige dem zweiten —
Bragg’schen Gesetze für die Reflexion der Röntgenstrahlen an Kris- tallen (Formel 2dsing = nA). Dass bei den Bragg’schen Versuchen die Reflexionsmaxima im Gegensatz zu den hier beobachteten aus- serordentlich scharf ausgebildet sind und bei anderen Einfallswinkeln kaum eine merkliche Reflexion stattfindet, ist dadurch bedingt, dass bei jenen Versuchen die einfallenden Strahlen eine scharf ausgebil- dete Periode hatten („homogene* oder „monokromatische“ Strahlen) und nicht nur zwei, sondern eine ungeheuer grosse Anzahl äquidi- stanter Netzebenen als Reflektoren für die Wellen dienten. Wegen der scharfen Form dieser Reilexionsmaxima konnte eine durch ihren Höhenunterschied bedingte merkbare Verschiebung ihrer Lage auch nicht stattfinden.
11. Nach der Kurve in Fig. 8 ist die Intensität des Reflexions- maximums erster Ordnung (2 = 1) nicht ganz zwei mal so gross wie die des Maximums zweiter Ordnung (x= 2). Wenn der Win- kel » wächst, nimmt die Anzahl. der von dem primären Strahlen- bündel getroffenen Elemente jedes Gitters (G, und G,) in umge- kehrtem Verhältnis zu sin g ab, und wenn das Reflexionsvermögen jedes Gitters in diesem Verhältnis abnehmen würde, so sollte aus
diesem Grunde das zweite Maximum der Kurve in Fig. 8 halb so ©
hoch sein wie das erste. Dass es in Wirklichkeit noch ein wenig höher ist, könnte dann dadurch erklärt werden, dass bei den Beob- achtungen nicht das ganze reflektierte Strahlenbündel, sondern nur der mittlere Teil desselben in den Empfängertrichter gelangte, wo- bei dieser Teil mit wachsenden Werten von & zunahm. Um die erwähnte Annahme über die Abhängigkeit des Reflexionsvermögen der Gitter von dem Werte des Winkels p zu prüfen, machte ich
“
A |
i+ py ye
ac an
_ ben worden waren. Es ergab sich, dass die am Gitter G, reflek-
REFLEXIONSVERSUCHE 25
| einige Versuche mit dem Gitter G, (Fig. 7) nachdem das zweite
Gitter (G,) entfernt und G, nach der Mitte der Scheibe AB verscho-
_ tierte Intensität für 9—= 69° 40 Proz. von derjenigen war, die dem
Werte » = 22° entsprach. Die von dem einfallenden Strahlenbündel
j getroffenen Teile der Oberfläche des Gitters verhalten sich in diesen - beiden Fällen wie sin 22° zu sin 69° oder 40,3 zu 100, welches mit
der Beobachtung vollständig übereinstimmt. Die am Gitter reflek- tierte Strahlung war also dem sinus des Winkelns » oder dem cosi- nus des Einfallswinkels umgekehrt proportional, was damit gleich-
bedeutend ist, dass beim Variieren des Einfallswinkels die reflek-
tierte Intensität der von den einfallenden Strahlen getroffenen An- zahl der Gitterelemente direkt proportional war. Es fragt sich, wie dies möglich ist, da ja bei wachsenden Einfallswinkeln die auf die einzelnen Gitterelemente fallenden Strahlenbündel immer enger wer- den (die Projektion des ringförmigen Resonatorkreises auf die Wel- lenebene der einfallenden Strahlen ist dem cosinus des Einfalls- winkelns proportional). Bei normaler Inzidenz wirkt nur der elek- trische Vektor auf die ringiérmigen Elemente. Bei schiefer Inzidenz
_ wirkt aber nicht nur der elektrische, sondern auch der magnetische
Vektor und zwar nimmt bei wachsendem Einfallswinkel die Wirkung
des letzteren zu, so dass die totale Wirkung lieber zu-als abnimmt.!
Nach den Bragg’schen Untersuchungen über die Reflexion der Röntgenstrahlen an Kristallen würde die Intensität der Reflexion z:ter
Ordnung für gleiche äquidistante Netzebenen nahe proportional !/r? sein, während Darwin,” der diese Frage theoretisch behandelt hat,
zu dem Schlusse kommt, dass die reflektierte Intensität proportional
1/7 sein sollte.
1 Durch Versuche mit einem ringförmigen Messresonator fand ich, dass wenn dieser Resonator sich in der Schwingungsebene der primären Wellen befand, so dass sowohl der elektrische wie der magnetische Vektor maximal auf ihn wirkte, die mit ihm gemessene Intensität etwas, aber nicht viel grösser war, als wenn der Resonatorkreis zur Wellenfläche parallel war, in welchem Falle nur der elektrische
- Vektor auf ihn wirkte.
2C.G. Darwin, Phil. Mag. April 1914, p. 689.
26 INTERFERENZ ELEKTROMAGNETISCHER WELLEN
Obwohl die letztere Folgerung den Ergebnissen meiner Versuche : recht gut entspricht, dari dies nicht als eine Bestätigung der Dar- win’schen Theorie betrachtet werden, da diese Theorie die Atome 4 der Kristalle keineswegs als ringförmige Resonatoren ansieht. Ewald,! | welcher auch das Reflexionsgesetz der Röntgenstrahlen teoretisch untersucht hat, nimmt allerdings an, dass die Atome sich wie schwingungsfähige Dipole verhalten. Aus gewissen Beobachtungen zieht er aber den Schluss, dass die Eigenperiode dieser Dipole — wenigstens bei gewissen Kristallen — gross gegen die der benutzten Wellen sei.” Es ist dies ein wesentlicher Unterschied zwischen den Versuchen über die Röntgenstrahlinterferenz in Kristallen und den oben beschriebenen in mehreren Beziehungen sonst analogen Ver- — suchen über die durch ein Raumgitter erzeugte Interferenz der Hertz’schen Wellen. Bei einer vollständigen theoretischen Erörterung dieser Versuche müsste auch — wie Ewald es für die Röntgen- strahlinterierenz getan hat — die gegenseitige Beeinflussung der zu ° den verschiedenen Netzebenen gehörigen Gitterelemente in Betracht gezogen werden.
V. ZUSAMMENFASSUNG.
Durch die oben beschriebenen Versuche ist eine durch ein dreidi- mensionales System von ringförmigen Resonatoren erzeugte Inter- ferenz der elektromagnetischen Wellen nachgewiesen worden, die in mehreren Beziehungen ein elektromagnetisches Analogon zu den Bragg’schen Versuchen über die Röntgenstrahlinterferenz in Kristal- len bildet.
Die Resonatoren waren in äquidistanten hintereinander stehenden Netzebenen in gleicher Weise angeordnet und hatten eine halbe Eigenwellenlänge von etwa 13,3 cm. Durch Versuche mit 26 cm langen Wellen (um eine genügend starke Reflexion zu erhalten,
1 P. E. Ewald Ann. d. Phys. 54 p. 519 (1917). 2 P. E. Ewald Physikalische Berichte, 2, H. 1. p. 59 (1921).
ZUSAMMENFASSUNG 27
, mussten die primären Wellen mit den Eigenschwingungen der Git-
_ terelemente angenähert isokron sein) ergaben sich zunächst die fol-
genden Gesetze für die Reflexion der Wellen an dem Raumgitter. 1) Einfallswinkel = Reflexionswinkel.
_ 2) Die Intensität der reflektierten Strahlung erreicht für gewisse Einfallswinkel Maximiwerte und zwar lassen sich diese Winkel (i) durch die Formel
2NA>C0S. U. — nah
darstellen, worin z kleine positive ganze Zahlen (die Ordnungsnum- mern der Reflexionsmaxima), d den Netzebenenabstand und 4 die Wellenlänge bezeichnen (geringfügige Differenzen zwischen den beobachteten und den nach dieser Formel berechneten Werten von i konnten auf einen Einfluss der mit wachsenden Einfallswinkeln folgenden Zunahme des Reflexionsvermögens der einzelnen Gitter zurückgeführt werden).
Diese Gesetze entsprechen vollständig den beiden ersten funda- mentalen Gesetzen, welche die Herren Bragg für die Reflexion der Röntgenstrahlen an Kristallen aufgestellt haben. Wegen der Dämpiung der elektrischen Schwingungen waren jedoch die mit den elek- trischen Wellen erhaltenen Reflexionsmaxima bei weitem nicht so scharf ausgebildet wie die mit monokromatischen Röntgenstrahlen erhaltenen. Zwischen den beiden beobachteten Reflexionsmaxima befand sich ein Minimum, dessen Lage durch die Formel 2 d cos i, —=(n--|) = bestimmt wird, wenn i, den diesem Minimum ent- sprechenden Einfallswinkel bezeichnet und z gleich 1 gesetzt wird.
Die Intensität eines Reflexionsmaximums war der Ordnungsnum- mer des Reflexes annähernd umgekehrt proportional, welches dadurch bedingt sein dürfte, dass —- wie durch besondere Versuche festge- stellt wurde — das Reflexionsvermögen der einzelnen ebenen Gitter dem cosinus des Einfallswinkels umgekehrt proportional oder, was damit gleichbedeutend ist, der Anzahl der von dem einfallenden Strahlenbündel getroffenen Gitterelemente direkt proportional war. ‘Nach den Bragg’schen Versuchen ist dagegen bekanntlich die Inten-
28 INTERFERENZ ELEKTROMAGNETISCHER WELLEN
sität des Röntgenstrahlreflexes dem Quadrate der Ordnungsnummer 7 proportional, während sie nach Darwin’s theoretischen Betrachtun- — gen der Ordnungsnummer selbst proportional sein sollte. Obwohl 3 die letztere Folgerung mit dem soeben erwähnten Ergebnis dieser Untersuchung im Einklang steht, darf dies nicht als eine Bestätigung der Darwin’schen Theorie betrachtet werden, da diese Theorie die
Atome der Kristalle keineswegs als ringförmige Resonatoren ansieht. Insofern man die Kristallatome als elektrische Dipole angesehen hat (Ewald), hat sich — wenigstens bei gewissen Kristallen — ihre Eigenperiode als gross gegen die Periode der Röntgenstrahlen (bezw. £-Strahlen) ergeben. Es ist dies ein wesentlicher Unter- schied zwischen den Versuchen über die Röntgenstrahlinterierenz in Kristallen und den in dieser Arbeit beschriebenen Versuchen. Durch Messung der an dem Resonatorensystem unter einem Kons- tanten und möglichst kleinen Einfalls- bezw. Reflexionswinkel (12°) reflektierten Strahlungsintensität ergab sich beim Variieren des Netze- benenabstandes (d), dass, wenn Oszillator, Messresonator und Git- terelemente angenähert isokron waren, der dem Reilexionsmaximum
entsprechende auf normale Inzidenz reduzierte Netzebenenabstand
(d cos 12°) angenähert gleich der halben Länge der einiallenden
Wellen war. Für eine Periode des Oszillators und des auf ihn abge- stimmten Messresonators, die kleiner als die Eigenperiode der Git- terelemente war, richtete sich dagegen der dem Reflexionsmaximum
entsprechende Netzebenenabstand vollständig nach der Eigenperiode
der Gitterelemente, und bei grösseren Perioden des Oszillators und des Messresonators war auch ein deutlicher Einfluss der Eigen- periode der Gitterelemente merkbar. Es konnte dies dadurch erklärt werden, dass die am Gitter reflektierte Strahlung sich im allgemeinen
aus zwei Teilen zusammensetzte, nämlich aus einer Eigenstrahlung
(Fluoreszenzstrahlung), die mit den Eigenschwingungen der Gitter- elemente isokron war, und einer zerstreuten oder erzwungenen Strahlung, deren Periode mit der der primären Wellen überein- stimmte (es bildete dies ein Analogon zu der von Barkla und Sadler
ZUSAMMENFASSUNG 29
4 nachgewiesenen Zusammensetzung der sekundären Röntgenstrahlung aus zwei entsprechenden Komponenten). Wenn die Periode der i primären Wellen kleiner als die der Gitterelemente war, dominierte ; die Eigenstrahlung des Gitters, während bei grösseren Perioden der _ Primärstrahlung die beiden Komponenten der reflektierten Strahlung sich geltend machten und eine Verschiebung des Reflexionsmaxi- mums nach der Seite der grösseren Werten von d hin herbeiführte. Durch Messung der von dem Resonatorensystem durchgelassenen mit den Eigenschwingungen seiner Elemente annähernd isokronen d Strahlung beim Variieren des Netzebenenabstandes (d) ergab sich _ eine Interferenzkurve, deren Maximum und Minimum mit Bezug auf ihre Lage mit dem Minimum bezw. Maximum der entsprechenden Reflexionskurve übereinstimmten. Einige derartige Durchlässigkeits- versuche mit einem dreidimensionalen Resonatorensystem hatte Ver- _fasser schon im Winter 1910 ausgeführt. Sie bildeten auch den Ausgangspunkt für die vorliegende Arbeit. Da für Perioden des Erregers und des Messresonators, die kleiner als die Eigenperiode der ringförmigen Gitterelemente waren, die am Gitter reflektierte Strahlung hauptsächlich aus der Eigenstrahlung des Gitters bestand und diese Eigenstrahlung weniger gedämpft war als die Eigenschwingungen des geradlinigen Messresonators, bot die Sekundärstrahlung des Gitters ein Mittel dar, stehende elektrische Wellen im Luftraume zu erzeugen, deren Länge von der Eigen- periode des Messresonators und der des primären Erregers unab- hängig war und sich nur nach der Eigenperiode der als sekundäre _ Emissionszentra wirkenden Gitterelemente richtete. Solche im Luit- raume objektiv vorkommende stehende elektrische Sekundärwellen konnten in der Tat durch einen Versuch nachgewiesen werden, bei dem die Eigenstrahlung des Gitters an einem ebenen Metallspiegel in normaler Richtung reflektiert wurde (objektiv im Luftraume vor- kommende stehende elektrische Primärwellen sind vom Verf. früher nachgewiesen worden.) Äbo, Physikalisches Institut der Akademie, im Februar 1921.
ACTA ACADEMIAE ABOENSIS MATHEMATICA ET PHYSICA I:4
ON THE
LITHOLOGY AND GEOLOGICAL STRUCTURE
OF THE
SIERRA DE UMANGO AREA,
PROVINCE OF LA RIOJA, ARGENTINE REPUBLIC.
A CONTRIBUTION TO THE KNOWLEDGE OF THE SUBANDEAN BORDER OF THE STRUCTURAL ELEMENT OF THE
„SIERRAS PAMPEANAS“.
BY
H. HAUSEN, PH. D.
ÅBO AKADEMI ÅBO 1921
we
u 5 iat
PREFACE.
Engaged as geologist of the Direction of Mines, Section of Geology, Buenos Aires, during the years 1914—17, I had to explore the geolo- gical structure. of a mountain complex, situated in the western part of the province of La Rioja, near the border of the Cordillera and named La Sierra de Umango. | started from Buenos Aires in beginning of may 1915 and returned in august the same year. Simultaneously with me a topographer from the Section of Topography, Direction of Mines, was committed with the mapping of the region. His map could, however, be used only at a time after my return to Buenos Aires, and later on I had no occasion to start a new expedition to this region with its very com- plicated geology for bringing the observations into closer accordance with the topographical map.
The geological sketch map, covering the reduced copy of the above named topographical sheet, accompanying this paper, expresses of course only the main geological features. In order to illustrate the intricate relations of the rock members of the area (mingled rocks, veins, banded ‘schists, etc.) a map of much greater scale would be necessary. The petro- graphical descriptions below may, however, give a picture of the litholo- gical relations, and that will be the chief purpose of the present paper.
The task of exploration of the Sierra de Umango area was undertaken on proposal of Dr Hans (Juan) Keidel, Director of the Section of Geology, Buenos Aires, in order to obtain a closer knowledge on the first hand of the relations between the structural element of the ,Sierras pampeanas“ to the east and the permian and andean (tertiary) structures to the west. That this matter of study really was to be found in the Sierra de Umango area, seemed probable already from an examination of the geological map of Ludwig Brackebusch (Gotha, 1891).
The geological mapping of the Sierra de Umango area should also be a continuation of the field works carried out in the near lying province -of Catamarca by the geologists of the Direction of Mines, Dr Walther
IV
Penck and Dr Juan Rassmuss, where ancient metamorphic rocks ana- — logous with them of the Sierra de Umango are the predominant.
The present paper is a somewhat modified edition or rather a working ‘ up in english of my original spanish report, addressed to the Direction of
Mines, Buenos Aires. The later will be printed in the Bulletin of the Direction (Serie B, Geologia).
I am much obliged to the Direction of Mines and especially to the ~ Director General, Engineer E. Hermitte, for the obtained permission to 3 publish this memoir in an independent way. ;
Academy of Åbo, Finland. September 1921.
CONTENTS.
Be GRAPH OF THE AREA No ds a a Nan dd
HISTORICAL REVIEW OF THE GEOLOGICAL EXPLORATION OF THE WESTERN-BART.OR TILECPROVINGE TA RIO JA dor ses sele 000:
PART I.
LITHOLOGY OF THE CRYSTALLINE, GRANITE INTRUDED BASEMENT.
A)
B)
C)
D)
E)
THE ,PREGRANITIC* SUPERCRUSTAL ROCKS OF THE SIERRA DE UMANGO AREA ....... USN Te LA ES RE MIRC RG ee Ann EEE [REG aumlinessedimentary, SCHISTSE u ur Se ee ota) ei b) The basic intrusive (and effusive?) rocks (metavolcanics)........... NRGoncordanksills.of amphibolitese an. as en PPA BEndotitichtock (si). sc N ae ee ae Syn Basıe,ikataclastic ‚schistsen. ea a le eye
THE FAMATINA GRANITE INTRUSION AND ITS EXOGENE PLUTO- NIERMEBAMORPEUSMSG er. Suse rl ayGraniter varities of they Pamatinay jee res. Seren b) Zone of abyssal mingling, brecciation and assimilation on the western
border olthey Kamatina batholith nn causes an Se je en@rosseutting) peamatiticdikesı breed er stalle u eee neh lese je ei se d) Plutonic metamorphism in the Sierra de Umango schist-body ......
AN EARLY PALAEOZOIC REGIONAL METAMORPHISM, SUBSE- QUENT TO THE FAMATINA GRANITE INTRUSION ................ a) Manner of intrusion of younger granites and syenites ............. b) Petrographical characteristics of the yonger intrusives ............
KATACLASTIC STRUCTURES OF THE BASEMENT ROCKS PRO- DUCED PRINCIPALLY BY THE LATER (TERTIARY) OROGENETIC IVT ONES ES NUTS Ele Ra vate pian Te bara ShA a bode le ta SL win SAL pM cro TEEN Aja neacrushed’: OTANUate di STATES! en as alls a ee falla ala H)E Deformations in the sedimentary schists 22). 006)... eles ale eee oe C)ueWataclasis, in: somessdanksschisise | Su a colitis er sie le se Else! se afen ee
GENERALITIES ON THE GEOLOGICAL POSITION OF THE SIERRA DE UMANGO AREA......... Apt SNS chon ORIG ER Cake iE cera Needs OF Rear
VI
PART II.
„POSTGRANITIC* CONTINENTAL DETRITIC FORMATIONS. VOLCANIC ROCKS. TERTIARY TECTONICS. GEOMORPHOLOGY, MINERAL
Page
OCCURRENCES ee 63 A) THE BASIS OF SEDIMENTATION OF THE CONTINENTAL STRATA 65 \ B) THE-CONTINENTAL-SERIES. 0 a u. N RA NA 66 a) Deseriptionok.localities 3... See SNES SRA 68 Bb) Summary ofthe stratigraphy; 2a 2 200 2 2. no ee ee 89 ©)’ THE; PERMIAN !OROGENETIG PHASE = 2.2. 2. 2.2.00 ee ee 96 D) VOLCANIC ROCKS IN THE CONTINENTAL SERIES... .... 2.002 73 a) Continental laya Tocks'. 22... Uk Sore ee ia) oes cee eee 98 å b) tuf layers... 5. ea SAG ee eee 102 3 E) CROSSCUTTING ERUPTIVE DIKES IN THE CRYSTALLINE BASEMENT 104 — F) 2 TERTIARY (2) VOLCANIC: ROGKS 120 0. Seo) N 107 @ G) THE TERTIARY AND QUATERNARY GEOMORPHOLOGICAL DEVE- LOPMEN Bo ola ioe cc's RE ke HN SSANG 109 a) \Avtirst) tertiary’ orogenetic’phase......8... u.a u ee 109 b) A levelling phase subsequent to the tertiary folding process. For- mation of a mountain gravel conglomerate ......................- 113 OT The ‘development of the main actual reliehy oss oo ee ee 116 d) The postpliocene erosion and accumulation............... BL: 121 1) A, first “erosion phase 320.45 «otro en RE N ra aja eee 123 2) Sheet@flood: erosion... << „une ce we ore eae ee 125 3) A later erosion phase. Linear cutting. River terraces.......... 127 4) Differences of the detailed sculpture in the crystalline rock-ground and in the terranes of the continental sediments ............... 130 HY MINERAL OCCURRENCES 2235000 ote. eee ie ee 132
LIST OF ILLUSTRATIONS.
PLATES:
l. Fig. 1. Mica-quartz schist. Fig. 2. Sericitic schist.
II. Fig. 1. Famatina granite. Fig. 2. Famatina granite, crushed.
II. Fig. 1. Amphibolite. Fig. 2. Biotite in transformation to
fibrolite.
IV. Fig. 1. Amphibolite schist. Amphi- Fig. 2. Quartzite with hornfels struc- bole partly transformed to ture (inclusion). biotite.
V. Fig. 1. Granite with dynamic flow Fig. 2. Aplite, completely granula- structure. ted.
VI. Fig. 1. Porphyrokataclastic schist. Fig. 2. Porphyrokataclastic schist. VI. Fig. 1. Kataclastic schist. Fig. 2. Kataclastic schist.
Per
Ed — .
24.
FASEN NS
Vil FIGURES.!
Page Key map comprising the middle part of the South American continent and showing the position of the Sierra de Umango area..................- 3 The Morro alto de Cordobés, Sierra de Umango ridge................. 7 Contorted limestone layers. NW from Carrizal ....................... 14 Pendelitie.sserpentumizeditockn... ya se skal Nols sheer aero a ely cart les BAR 28 Sketch map, comprising the central part of the Famatina batholith, sho- wing the palaeozoic schist inclusions (after Bodenbender) .......... 30 Great individuals of red potash feldspar. Pegmatite NE from Villa Union 41 Profile along the western side of the Cerro Guandacol, showing the Gond- WeaKamlaviens..) Sketch from the Matures nz das. u ne 69 Profile through the lower Gondwana, Quebrada de los Lagares, Cerro Valle ae Na yt Sa SR Pe NEN 73 Panorama view to the south of the eastern side of the Cerro Villa Union 74 The Gondwana strata in the Cerro Punta Colorada. Looking south..... 75 An erosion remnant of red Gondwana sandstone. Las Rarmaditas ...... 76 A part of the ,intermontane sandstone zone“. Looking SE............ 79 Tertiary arenaceous strata. Northern part of the ,intermontane zone”. [EO GKINGESO UCISSS 555 21. vat I RE I ES a SVs (2 aR An PES ler a 82 Trestillire occursence, at AouarBlanguttan ee ne ae een. 90 Lava and tuff accumulations of tertiary age. North from Tambillos.... 108 A part of the „intermontane zone‘, showing bending of the strata. Loo- Ba INNE 0 oj EN LE ea RER ee Nee a 110 Tectonical profile (E-W) from the western side of the Sierra de Umango nozer showing twolold dislocalonsah urn en En tol oly risks fe) ol -111 Tectonical profile (E-W) irom the same part showing an overthrust phe- HOIIKEITOT ER LAN 2 RN ee else Male Aa od ehe 112 Overthrust folding at the Cerro La Pampa. Eastern border of the „inter- LITO MAG TGS ER En ae AR ee ee LI? Sunken remnant of a young tertiary conglomerate. Crest of the Sierra deslimansonridgert Booking) south aan een. ee 114 Tectonical profile (E-W) from the eastern border of the „intermontane zone“ showing dislocation of quaternary gravels ....................-- 124 High destructive diluvial terraces at the slope of the Nevado de Fama- [LINA SEI Ga buerfo,Nlegrer - u fees ala I ans I Se neigen nee 126 Eastern side of the Sierra de Umango ridge showing young erosion in SHEEINIIOOGN surfacesı Sia cit A en en ne ie. 128 Highland topography near the crest of the Sierra de Umango ridge. Loo- NEE NER is, « (öl Tre chapeau te epee Mirage Lave ene ha B/S oi Sisal ais avs Bree ee 131
1 All the landscape views and profiles sketched by the anthor.
Fo “Ee >
PHYSIOGRAPHY OF THE AREA.
Sierra de Umango is the general name of a mountain group,
situated in the deep corner between the eastern border of the Cor- dillera in the west, the southern edge of the Puna de Atacama in
the north and the lofty, meridionally streching ridge of the Sierra
eS wy, ee
(and Nevado) de Famatina in the east. It lies in the western part
of the province of La Rioja, in the extremely dry, so called suban-
_ dean zone of the central western Argentina. The closer position of the _ area is defined by the coordinates 28° 50’—29° 50’ S Lat. and 68° | -00’—69° 00’ W Long. from Greenwich. The geological study in this | region comprises, however, more than the topographical unity of the Sierra de Umango and extends sometimes to the foothills of
N 7 > As &
A U A a
= A an
u all iS Pt
the surrounding mountains. The approximate limits of the surveyed
area are thence: in the west: the eastern border of the Cordilleran block, or nearer said the „Precordillera“, marked by the steep
western wall of the tectonical valley Guandacol — Tambillos —
_ Guandacolinos, and further to the north by the western border of the
Bolson de Jaguel; in the north by the southern edge of the Puna de Atacama; in the east by the western slope of the Sierra (Nevado)
de Famatina; in the south the southern limit of the alluvial plain of Pagancillo and the water gap of the little river Vinchina, above
its confluence with the Rio Bermejo. (See the accompanying map!)
Regarding the orography, two elements may be distinguished:
1) the alpine mountains of the Sierra de Umango in proper sense,
the Cerro Cacho and the Cerro Villa Union, all comprising the old
_ erystalline rock ground, and 2) the surrounding foothills with the
2 SIERRA DE UMANGO
lowland ridges, built up by soft, younger rocks, principally of
clastic origin. Of the lofty „erystalline“ mountains, the Cerro Cacho lies nearest to the Cordilleran border and is for a distance of some
miles closely united with the later (Pampa de Leoncito). Separated ;
N ;
a
from the Cerro Cacho by a relatively hilly highland surface there © follows to the east the very ridge of the Sierra de Umango, run- | ning NE--SW, with a somewhat lower and shorter parallel chain : to the east, the Cerro La Pampa. To the SE off these ridges lies i
the more isolated complex of the Cerro Villa Union, on all sides ©
surrounded by alluvial deposits.
All the named mountains are of quite considerable heights, as -
shown by the following data:
Gerro(Gachoerwas oe Fae 4,660—4,405 m ab. the sea. Sierra de Umango ridge ... 4,300—4,275 , , , 5 Cerro Villa Union. 8.2... ISDN RN x
The absolute heights of the valleys vary between 910 and 2,200 —
m. The foothills and the lowland ridges of softer rocks and also ©
isolated groups of the old basement rocks (Villa Castelli) attain only some hundred meters over the valley floors. Of greater importance
among the foothills are the Cerros de San Antonio to the S from
the Sierra de Umango ridge, the Cerro Bola SSE from the Cerro Villa Union, the Cerro Colorado, lying immediately to the east of
the Cerro Villa Union and at last the broad ridge, well marked by —
a cuesta to the E, forming the dividing line between the Jaguel -
bolson and the Vinchina valley in the direction NE from the Sierra de Umango. (See the topographical map!)
The whole western part of the province of La Rioja belongs to the climatic province of broader sense called by G. Davis! ,Za zona andina“. This is, as known, characterized as a dry desert, where rainfall occasionally occurs only in summertime, when the Eastwind blows (from the pampa). In the higher mountains a western storm
1 El clima de la Republica Argentina. Buenos Aires. 1909.
PHYSIOGRAPHY 3
Fralled „la zonda“ is thundering on nearly the whole year. This wind is cold and dry and originated probably by a great, permanent & minimum, extending over the pampean plain further eastward.
j The summits of the Umango Mountains are never snow clad. É Ey the giants outside the area, reaching up to more than 6,000 om, such as the volcan Cerro Bonete on the Puna-edge and the eae de Famatina, carry a white cap. The snowfall here is more
x
abundant during the summer than the dry wintertime.
+
eql*A Sep
We
Zhe R Camaroons Te
x SAT: gous
o RTıgte > NA ZEHN:
NOR "Cape hrio
RR s.L’Salto Guaira So, ie eS TREE Jp NA Mae SEE Ne OS SR SO Tropic * oF "Capricorn
ER
a SIE lan EU I Tevet
i Sea Merges inet
= =
= ge =
SS \ (Oman.
7
x
a a AS TASTE oe
5. ee
E > EU a “Bile fros hte Pista 6 m "Tand ENGLISH MILES
aan AR o N de SN > . : © 100 200 300 400 500 cn ÄR SS ss ifs jes A é Ves > cs "2 Blanca none 3
Fig. 1. Key Map. @ The Sierra de Umango Area.
A peculiar scenery of these deserts are the heavy skyfalls in the moist season, the so called „crecientes“. When such an outburst
Pos eee
HX
Er
: occurs on a mountain slope, all the dry „quebradas“ or gorges are 4 filled by rushing streams, loaded with boulders and gravels, sand 4 and mud, cutting through old piedmont deposits, filling preexistent Ki
channels and forming new ones in the broad alluvial fans, radiating
4 3
R
4 SIERRA DE UMANGO
from the mouth of every „quebrada“ into the valley floor outside i the mountains. The rugged topography of the area, often a veri- table „Bad Land“, is doubtless the result of the intense, but spora- 3
dic work of the named „crecientes“.
The day-temperature is generally high all the year round. During
the wintermonths the thermometer often reaches 20° C, resulting from
the intense insolation. The nights are cold with 10° and more ©
below the zero. On heights ab. 4,000 m the temperature is generally low even at daytime, because the icy ,zonda“ never ceases to blow.
Running waters are, except the sporadical ,crecientes“, extremely scarce in the Sierra de Umango area. Over the sandy bottom of the Guandacol valley the little Rio Bermejo is meandering southwards to the bolsons in the province of San Juan (see below, page 120) farther south. This river receives from the left hand an
insignificant tributary, the Rio Vinchina, coming from the broad - valley, which separates the Sierra de Umango from the Nevado de -
Famatina. Its upper course is draining the Jolson of Jaguel through
the water gap opposite the Vinchina settlement. Both rivers are
speeded by underground waters, and a source can hardly be fixed. All the mountain valleys and gorges are dry.
A continuous vegetation is never to be seen, except the ,pasto- 1
rales on great heights, where a hard grass grows, serving as food to — the cattle. The ,barrancas“ in the dry valley trains are sometimes °
bordered by algarrobo trees („gallery woods“), indicating the course
of the underground waters. In the immediate vicinity of the water
springs, generally scarce, and occuring either at the foot of a moun-
tain or on the bottom of a ,quebrada“, there grows a high grass, —
called by the natives „carrizales“. The open plains of alluvial gra- vel, sand and loess are dotted by xerophil thorn-bushes and various
kinds of cactus. The most characteristic form of the northwest- -
argentine deserts, the giant cactus (cereus giganteus), is completely —
lacking in the Sierra de Umango area; it does not extend westward _
from the Nevado de Famatina.
Sir Od ey eee ar
PHYSIOGRAPHY 5
The settlements are not frequent, especially the villages. These exist on irrigation culture. The grape cultivating Villa Union and the Villa Vinchina are the most important, both consisting of claybuilt block houses, forming rectangular ,quadras“, separated by muddy Streets. Outside the dwellingplaces there extend the vineyards, pastures and a network of very primitive irrigation channels. Both villages receive their water supply from the Rio Vinchina.
In the mountains there are several small „puestos“ or „pircas“, stone dwellings, inhabited by „gauchos“, a poor halfblood, watching sheeps and cattle.
The region lies quite outside the lines of communication and far away from greater centres of population of the Republic. The nearest railway station is Chilecito, on the eastern side of the Nevado de Famatina, the end of the Serrezuela branch of the Cérdoba— La Quiaca railway (Ferrocarril del Norte Argentino). The way from _ Chilecito to Villa Union is a riding path over a quite rough ground and running across the southern, lower part of the Nevado de Famatina. The freights from and to Villa Union are carried on mule- and donkey-back.
HISTORICAL REVIEW OF THE GEOLOGICAL EXPLO- °
RATION OF THE WESTERN PART OF THE PROVINCE LA RIOJA.
In the mining history of the Argentine Republic the province of
La Rioja plays a very important part. Already since times of the spanish jesuits work has been done in these distant and desertous . mountains for the search and exploitation of many rich minerals of gold, copper and silver. Especially in the famous Nevado de Fama- tina the mining work has been very productive. A rational exploi- tation of the ores there began, however, about fifty years ago, and
since this time the mining industry here has been in a steady pro-
gress. The newly built railway from Serrezuela (Prov. de Cordoba) to Chilecito (situated on the eastern side of the Nevado de Fama- tina) and the electric wire rail from Chilecito up to the mines of La Mejicana, in an alture of about 5000 m. (ab. the sea) on the E slopes of the Nevado, give an evidence of the intensity of exploi- tation in the last times. At present (1917) the whole business is, in consequence of various circumstances, stopped.!
In comparison with the extensive prospecting work of all the re
tal occurrences in these regions, the progress of the geological ex- ploration has been quite insignificant, especially when regarding the old crystalline rocks, appearing as the main components of the mountain blocks of the Sierra de Velazco, the Sierra de Famatina and the Sierra de Umango.
1 Regarding the mining history of the Nevado de Famatina, see: A. Stelzner (1876--1885), and E. Hermitte: La mineria argentina. Buenos Aires 1915.
3 KSSS vun m Pr v Pe ee ee = Fa RE, ee eae tn oe ee ce Th CO MS LS ET. 70 oh pn EE rind eae
EC ee ree ee ee ee ee TS
Bg clare pert nn
Y
i fs +
b S ; ;
Dr A. Stelzner (1876—85)! a late professor of geology at the Uni-
4 versity of Cördoba. During the summer 1871—1872 he traversed the province of Catamarca (in the N) and arrived also southward to the
Nevado de Famatina with the purpose of making a slight investi- gation in all the mining fields here. During this trip Stelzner
made an important discovery. In Potrero de los Angulos, on the
Fig. 2. The Morro alto de Cordobés (about 4300 m.), Sierra de Umango ridge, seen from the south. The eastward dipping lines indicate the zone of lustering mica schists (see pag. 17).
eastern side of the mountain he found fossiliferous schists of silurian age. The fossils were later described by E. Kayser (see later on). In his report Stelzner says, that the Nevado de Famatina differs from the so called ,Sierras pampeanas“ by the absence of gneissic rocks and by the predominance of slaty and quarzitic schists, further by more clastic members, having in common only the large granitic bosses.
1 Compare the bibliography.
8 SIERRA DE UMANGO
A second expedition of Stelzner, realized during the following summer, was directed to regions more to the south, into the pro- vince of San Juan. During this journey Stelzner’s object of study | was principally the continental, clastic formations, reposing with creat | unconformity on the metamorphic schists. He discovered a rhaetic |
flora, since closer examined by Geinitz (1876).
The next geologist, who visited these regions, was Dr L. Bracke- 3 busch, formerly professor of geology at Cordoba too. During extensive © and audacious journeys, this infatigable explorer twice crossed the °
region of the Sierra de Umango. In his „Mapa geolögico del inte- rior de la Republica Argentina“ + he has fixed, according to his and Stelzners observations, the main geological contours of these regi- ons. Brackebusch divided the formations here as follows: 1) ar-
chaean, 2) rhaetic, 3) tertiary and quaternary. The whole material ©
collected by Brackebusch was then handed over to various spe-
cialists in Germany, where also all the results were published. As x
these laboratory-investigations generally were carried out without any :
geological basis, their value is only a descriptive mineralogical one _
(resp. palaeontological). It is a pity, that Brackebusch never got any occasion to publish all his own extensive impressions except what „Mapa geolögico del interior de la Repüblica Argentina“ contains.
During the later years Dr W. Bodenbender, at present professor of geology at Cördoba, has carried out a more systematic regional investigation in the western provinces, partly in the subandean region of La Rioja (1911). The Sierra de Umango area was, however, visited by him only to a small extent, principally along the borders of the Cerro Villa Union. But his exploration of the southern part of the province of La Rioja and also his monographical treatment of the Nevado de Famatina (1916), Sierra de Umango’s neighbour to the E, has offered
much of comparative material to the study of the Sierra de Umango-
geology.
1 Scale 1: 1,000,000. Gotha 1891.
AF
HISTORICAL REVIEW ' 9
In still later time Dr Walther Penck, formerly geologist of the
Direction of Mines, at present professor of geology at the University of
- Leipzig, Germany, has investigated the area of the Bolson de Fiam- _ bala and surrounding mountains, lying in the province of Catamarca _ and in the NE continuation of the Sierra de Umango complex. The
results of these works are: a geological map in the scale 1: 200,000, preliminary reports (1914/15) and a complete geological-morpholo-
_ gical monography (1920) of „the southern border of the Puna de _ Atacama“, as he names the region.
The main results of W. Penck may not be mentioned here, because in the following pages it will be various occasions to return to his works. It may only be stated here, that the peculiar ban- ded structure of the Sierra de Umango metamorphic series is
continuing far into the Catamarca mountains and that the conti-
nental overlapping strata show a great similarity to the La Rioja beds. Simultaneously with my own camp works in the Sierra de Umango
area, my colleague Dr Juan Rassmuss was carrying on an inves-
tigation in the adjacent part of the „Precordillera“ to the W and
SW from the Sierra de Umango. The result of Rassmuss is _ hithertoo, so far as I know, a handcolored geological map. His verbal
communications have been to great use for my own purposes.
DakıL
a LITHOLOGY OF THE CRYSTALLINE, GRANITE. > ti _ INTRUDED BASEMENT. ° \
A) THE ”PREGRANITIC” SUPERCRUSTAL ROCKS OF THE SIERRA DE UMANGO AREA. a) THE SEDIMENTARY SCHISTS.
Engaged in a careful study of the lithological composition of the Sierra de Umango crystalline basement, one must arrive at the
conclusion, that there is a great bulk of sedimentary and effusive
members, — a „pregranitic“ unity, while granites older than the
postsilurian(?) Famatina granite, do not occur there. The sedimen- tary part of this bulk consists principally of three distinct lithological
groups: limestones, quartzitic schists and argillaceous schists. The relative quantities of these components are very difficult to
estimate, because the original succession is lost by various proces- ses of disturbances, causing faults, folds, charriages (overthrusts),
etc. with accompanying thickening and inversion of the strata. The
limestone-amphibolite banks with their fantastic torsion pheno- mena offer an evident proof of the high grade of disturbances,
suffered by the basement in question (see the fig. 3).
A conspicuous circumstance is the great thickness of the limestone banks especially in the western part of the region, where the lime-
stone alone forms whole ridges and hills, the total thickness of the
ayers attaining some hundred meters without any tectonical thicke-
ning. This abundance of limestone stands in a marked contrast to the experiences in the Famatina mountain, where such layers are
thin and scarce!. Also in the other pampean sierras we find
1 See: G. Bodenbender (1916).
14 SIERRA DE UMANGO
limestone to quite an insignificant extent, except in some parts of 7
the Sierra de Cördoba and in the Sierra de la Huerta. Among the other sedimentary members in the Umango pregra-
nitic series the micaschists occupy different, broad zones, the most i important of them being a more central one. The quartzitic schists
are always intercalated in these rocks, partly in regular alternation, partly in mightier zones.
Fig. 3. Contorted and thickened layers of silicious limestone (white) in micaschist (grey). NW from Carrizal. Height of the rock-wall about 50 m.
Apart from these principal types to a smaller extent we find
interstratified members of strongly pressed, dark schists. These may be considered partly of sedimentary, partly of pyroclastic ori- gin.
All these layers form a concordant prism without any sign of break in the sedimentation. Conglomeratic layers are completely lacking.
Below follows a closer lithological characteristic of the named sedimentary types. |
1 A. Stelzner: Mineralogische Beobachtungen im Gebiete der Argentinischen Republik. Tscherm. Min. und Petrograph. Mittheilungen. IV. 1873. Pag. 230, and foll.
en NT u en ee She de ne "re ee ee es re
PE ee SS
Ada FaR RR
SIN
a De ara er en
circumstance has caused, that a skarn (silicatic) mineralization very
4
CRYSTALLINE BASEMENT 15
4 ‘The limestone is characterized by a high grade of purity. This ‘seldom occurs, in spite of the abundant, crossing granitic veins. In some parts there may be found a hornstone, a characteristic mark of ‘the precordilleran silurian limestone, and formed probably from remnants of silica-bearing organisms (radiolaria). The mightier layers
of the limestone are generally divided into thinner banks by joint
; planes, seldom by intercalations of foreign matter. The banking of the rock marks very well the folding and torsion phenomena.
In handspecimen the rock shows generally an evengrained structure
- (sugargrained) and the colour is mostly pure white. In some parts there are very excellent kinds of marble. An analysis of the lime- stone of usual habit from the Cerro Potrero Viejo, near to the
Puesto del Agua, was obtained from the Chemical Laboratory
of the Ministery of Agriculture, Section Mineral Waters, Buenos
Aires: SION A ESRI UME at tay 4,65 /o BO A N ER 41,09 °/o OE (OPT le Ae N N 35,30 °/o TRAN GH Ga a pena aia ee 18,04 2/0 NE OA OUI N la ö SAO VN SA ates ra ESS | hou a
Total 98,21 °/o
From this analysis it is evident that the rock contains a consi- derable amount of MgO, but not to such a degree, as to be named dolomite.
Other crystalline limestones from the area of the pampean Sierras, i. e. the quarry Rio del Sauce, Sierra de Cordoba, whose analysis-data were communicated by my colleague Dr R. Beder, show scarcely more than traces of MgO in eleven limestones ana- lysed. Over all the Cördoba-limestones (Malagiienio, Santo Do-
16 SIERRA DE UMANGO
mingo, Alta Gracia, Mal Paso, etc.) are very poor in MgO.! The | |
Cördoba limestones are considered of lower silurian age.?
The limestones of the Precordillera, i. e. of silurian age too, but ‘ unmetamorphosed, from Salagasta, province of Mendoza, show very 5
small amounts of MgO.
Microscopically the Umango limestone shows a structure of isometric
grains, with quite angulated contours. The characteristic twinning-
lamellae after — 1/2 R are shown elsewhere. Accessoric minerals are met =
with only at a small extent and consist principally of quartz and sericite. A dominant direction in the structure is lacking and also every sign of kataclasis.
In parts where the limestone occurs only as thin interstrati- fications in quartzitic schists etc., it has a more grayish-bluish colour and shows almost a banded transition to the neighbouring rock.
A common feature is a banded intercalation of basic effusive layers (amphibolitic schists) in the limestone, a peculiarity treated off later on.
The quartzitic schists occur, as stated above, partly as broader, partly as thinner zones or interstratifications. In the first case more schistose types of grey and dark colours are met with on the - eastern slope of the Sierra de Umango ridge. In the second case there are milkwhite, glassy layers or more concordant veins. The quartzitic schists have the schistosity planes marked by mica (biotite), often leading to quartzitic mica schists. The milkwhite quartz veins of layer-appearance occur principally in mica schists, where there are also thin layers of limestone. The intimate alter- nance between these layers may produce a marked banding struc- ture on the rock surfaces (comp. later on).
1 A. Cogliati: Consideraciones sobre algunos calcareos y cementos de la Republica Argentina. Buenos Aires.
2 H. Keidel (1914), page 674, and R. Beder: Las cales cristalino-granulosas de la Sierra de Cordoba y sus fendmenos de contacto. Direcciön G:ral de Minas, Geol. e Hidrolog. Boletin num. 7. Serie B. Buenos Aires 1914.
og
Bug a a a ET ERE en ar er ee Br ae De een
ES eS ne oe a Yan
CRYSTALLINE BASEMENT 17
On the transverse fracture surfaces of the very quartzitic schists may be observed too quite a distinct glassy habit. These rocks are, as shown later on, strongly recrystallized. The mica marks the very bedding planes of the rock (Pl. I, fig. 1), as becomes evident from the fact, that these schists partly carry interstratifications of sericitic schists. With regard to the relatively considerable amount of mica in the mass of the schists, it is quite probable, that the schist in its actual metamorphic habit represents an original psammitic rock of the arcose type, where the feldspar has contributed to the forma- tion of mica. The mica is in most cases a biotite, oriented in subparallel direction between the quartz grains. Other varieties of he schists contain also sericite, whose relative quantity rises occa- sionally forming sericitic schists. Garnet of reddish colour also occurs, sometimes quite abundantly, with rounded contours and of
- granoblastic habit. The whole microstructure indicates a strong recrystallisation, undergone by the rock.
_ The mica schists occupy, as mentioned, principally the broader zones, relatively free from granitic intrusion, but very intensely pe- netrated with concordant sheets of amphibolite. The most important of these zones runs over the Cerro Villa Union and over the Sierra
de Umango ridge with a general NNW-trend.! Another zone with
‚heavy amphibolitic sheets occurs in the Vinchina valley ground, exposed on the surface in the Asperecito hills close to Villa Castelli.
The ordinary habit of these schists resembles the alpine ,,schistes- lustrées“ with silverglittering planes carrying abundantly quartz-lenses and garnet aggregations. The schistosity planes are uneven, generally
tippled. The amphibolitic layers are strongly concordant in the rock structure. The mica is generally represented by muscovite, but in many parts biotite also occurs, attaining the same relative quan- tity as muscovite. The last case results in a type, corresponding to a two-mica-schist. Another variety is a sericitic schist of pale greenish colour and with intensely lustering planes. It is garnet- Serowsee fig. 2, pag. 7.
18 | SIERRA DE UMANGO
bearing too and very intensely microfolded. The occurrence of this | type is quite limited, appearing, as mentioned, only in alternance — with quartzitic schists. Microscopically the sericitic schist shows a considerable amount of ferri-oxyde, forming parallel thin zones and q very markedly indicating the microfold-structure (see Plate I, fig. 2). 4
The mica schists are also to be found in various parts of the Umango area as small intercalations between other thin layers in the so called banded rocks, below more closely characterized. The 4 highly granitized areas, as the summit of the Cerro Cacho and in — the eastern part of the region, are also represented by mica schists. a The mingled product, called by Sederholm migmatite, and also j wholly resorbed rocks, appear many times here, as e. g. in the q western foothills of the Nevado de Famatina. 4
According to Bodenbender there are abundantly in the last 3 named mountain in the higher regions, wholly surrounded by gra- q
nite (ci. the little sketch map, page 30), micaschists with basic in- ~
trusive sheets of exactly the same appearance as in the Umango ~ area. They represent metamorphosed, cambrosilurian slates. The — writer had occasion to see these schists, as he crossed the moun- q tain on his return journey from the Umango district, and he could — prove their similar appearance. |
Whereas in the Famatina area there are in some parts but slightly — metamorphosed facies of the slates and other sedimentary rocks - (the fossiliferous parts), all the sediments of pregranitic age in the ° Umango area have a very strong metamorphic habit to make it - hopeless to find any tracks of fossils or even to meet any possi- 1
bility of studying the original lithological structures. The changes —
have been not only of contactmetamorphic nature but also kata- clastic in many ways.
If it is correct to consider the easternmost part of the Umango | complex, situated nearest to the Famatina granite batholith, as the a lowermost horizon, and the western parts generally as crustal masses of a higher position, one may state, with reference to the distribution
CRYSTALLINE BASEMENT 19
4 of the sedimentary types, that the quartzitic and mica schists occupy
Pe. > =
: the lower, the great limestone banks the higher stratigraphical levels. ‘a
E72
The question about the relations of the Umango sedimentary series to the geological structure of the surrounding countries and the problem of the age of these rocks will be treated of below.
Discussion of the Geology of the Umango „pregranitic“ sedimen- tary pile. |
As seen, the series are composed principally of three types
of rocks, representing the three principal members of products
of subaerial decomposition by chemical way: lime, silex and argile,
with some transition products especially between the two later. It may be presumed, that the basement of the sedimentation was gra-
dually sinking, because the limestone banks occur in the higher _ horizons. The continental surface, from where the decomposed
material derived, was probably an ancient one, composed of acid
‘silicate rocks, as for instance the old precambrian terranes. It is
_ probable from the heavy masses of quartz-sediments in the Umango
area, that many acid rocks really occurred in this terrane. The complete lacking of conglomeratic layers in the Umango series and also of other traces of stratigraphical breaks makes it very probable, that the sedimentation proceeded quite uniformly.
Above was stated, that in consequence of the intense metamor- phic changes in the Umango series it is impossible to determine the age of the rocks by way of fossils. One will arrive at a defi- nite conclusion with this question by comparing the Umango rocks with others occurring in the neighbourhood.
Above it was already stated, that the schists of the Umango area show a great resemblance with the similar metamorphic schists oi the Nevado de Famatina. Whilst the last are considered by Bodenbender of cambrosilurian age and are intruded by the same granite (the Famatina batholith), both in the Umango and in the
- Famatina, and older granites do not appear in the Umango, it seems
20 SIERRA DE UMANGO
very probable, that the Umango schists are homotaxial with the 4 |
Famatina schists. | |
The great limestone banks in the Umango area are met with principally on the western side, in the Cerro Cacho and on the lower mountains between her and the Sierra de Umango ridge. By studying the map recently worked out by Rassmuss illustrating the Precordillera of the province of La Rioja, one becomes convin- ced, that the Umango limestones once formed the northward con- tinuation of the mighty lower silurian zones of the said mountain chain. Therefore, the Umango limestones may be of lower silurian age too. The circumstance, that the later are strongly metamor- phosed and contorted, whilst the limestone masses of the Precor-
dillera are unmetamorphosed, does not form any obstacle to the presumption, and may be explained by the possibility, that the
Umango is forming another structural element in respect to the Precordillera.
Studying the report and map of W. Penck (1920) from the wes- = =
fern part of the province of Catamarca, to the north and northeast from the Umango area, one will find that the intensely folded, banded and, metamorphosed structure of the Sierra de Umango
continues NE- and NNE-ward under the floor of the Jaguel bol-
son far into the mountains of the western part of Catamarca, an area lying to the west of the valley of Fiambalä. To the east the Catamarca structure is bounded by a granite area, corresponding to the Famatina batholith (which also forms the eastern margin of the Umango structure). The Catamarca structure may be of lower pa- laeozoic age! and undoubtedly it is of the same as the Sierra-de- Umango-element. The limestone is there, however, completely lacking.
1 Penck has expressed the opinion, that it is of devonian age, but while fos- sils are also here lacking, this statement is of no definite character. Nevertheless, no doubt, it is palaeozoic and of pre-Gondwana-age at least (see later on page 61).
Vv N \ x \ Te, a ee ee ee ee Ege, ee
a um zn
CRYSTALLINE BASEMENT 21
The Precambrian Basement.
| Above we have presumed, that the sedimentary members of the a Umango structure are derived from an old surface, composed pro- | bably of chiefly acid crystalline rocks, i. e., a precambrian basement. The next question which arises, is where to find this basement, and ii it still exists in this part of the Republic.
The Umango schists are rooting in an irruptive granitic mass, the Famatina batholith. Everywhere to the east the schists are boun- ded by this granite with its many dikes and embayments. Thence a precambrian complex is not visible here. Going farther to the east, into the Famatina mountains, one meets with complexes of the mentioned cambrosilurian schists, on all sides surrounded by the irruptive granite (see the sketch map, page 30). Here also an older element is lacking. Bodenbender mentions, that the Fa-
_ matina granite extends eastward to the very foot of the Sierra de
Velazco, a mountain mass rising to the east of the Chilecito valley. On the western slope of this mountain an old gneissic rock extends in a northerly direction as already recognized by Stelzner and Brackebusch, and fixed on the map of the latter. The gneiss shows an itruptive contact against the Famatina granite. On the crest of this Sierra we find again the same younger granite (see the corres- ponding profile of Stelzner), probably only a branch of the great Famatina intrusive body. This old gneiss may probably be regarded as the precambrian basement. A confirmation of this statement will be found in the relations further to the north, in regions of the province of Catamarca. Here is found a mountain system to the east of the above named valley of Fiambalä, more closely inves- tigated by W. Penck. This terrane consists of old schists and gneissic rocks with younger granitic intrusions as in the Sierra de Velazco. Between the schists-gneisses and the western element, i. e., the continuation of the Umango structure, there occurs a large gra- nitic body, according to Penck, belonging to the Famatina batholith.!
1 See the map of W. Penck (1920).
22 SIERRA DE UMANGO
The old rocks to the east of the valley of Fiambalå are again divided further to the east by granite from an extensive, precambrian terrane in the Sierra del Aconquija mountain complex, studied by Rassmuss. This last named precambrian shows to the east a marked unconformity against a still older gneissic complex, regarded by Rassmuss (1916) as archaean.
The great Famatina—Fiambalä intrusive body of lower palaeozoic age seems to form a kind of interformational batholith between the Umango structure to the west and the precambrian gneisses to the east. One may assume, that the Famatina intrusion was cutting off the connections between the lower palaeozoic structure and its base- ment, so that an unconformity between these two elements nay probably never be found here.
b) THE BASIC INTRUSIVE (AND EFFUSIVE?) ROCKS (METAVOLCANICS).
The very abundant intercalations of basic schistose rocks between the doubtless sedimentary schists are characteristic members of the „pregranitic“ schist complex of the Umango area too. These basic rocks are partly ordinary amphibolites, according to the definition given by Rosenbusch, partly of different aspect and almost diifi- cult to be classified, because of the intense kataclasis undergone by the rocks. But it may be assumed, that the greater part of these dark schists represents igneous (partly gabbroid) injections into the sedi- mentary schist masses. These injections appeared already at a time, when the sediments still lay in their original position, i. e., before the first palaeozoic diastrophic movements and before the intrusion of the Famatina batholith took place. The basic rocks are therefore „pregranitic“ too.
The original relations between the sedimentary schists and the basic sills have been destroyed at different times to a considerable extent. The thin layers and lenses of amphibolite in the limestone masses are certainly in most cases only the remnants of thicker volcanic sheets, afterwards, during the folding movements, extremely
fr
CRYSTALLINE BASEMENT 23
thinned out and torn. The banded appearance of the amphibolites in the granitic injection zones of the more central parts of the area,
o— as will be demonstrated later on — a very common phenomenon,
is produced principally by an exfoliation-process of thicker ampbi- bolite layers by way of concordant granitic injection. Only in the zones of the „schistes lustrées“ the amphibolite sills occur in a more preserved stage, a fact due principally to the almost complete ab- sence of granite veins. An exception from ihis rule is to be found in the hills of the Cerro Asperecito, the Cerro Pintado and the Cerro Tupe in the Vinchina valley, where the amphibolite layers are intensely penetrated as to form an arterite (Sederholm), due to the imme- diate vicinity of the Famatina batholith. But the appearance as a whole, the concordant alternance between the schists and the sills, is still conserved here too.
On studying the microstructure of the basic rocks, it is seen, that hardly in anyone part of the area a primary or an almost primary one occurs. The structure is either a kataclastic or a recrystallized
one, because undergone all diastrophic movements of different times.
The basic rocks may here be divided into three groups, as follows: 1) Concordant sills of amphibolites. 2) A peridotitic sill rock (local occurrence). 3) Basic, kataclastic schists of doubtful origin. Tho these belongs still an isolated type of porphyritic character and of local occurrence too. While the rock is of less basic character, it will be described apart from the enumerated types.
1) Concordant sills of amphibolites.
The greater part of these amphibolites occurs, as mentioned,
as mighty, alternating sills in the mica schist zones, where they
often appear with astonishing regularity. They are also found in the quartzitic schist masses and in the limestone banks, where the Sills are generally thinner. The contorted appearance of the Jimes- tone has already been mentioned.
24 | SIERRA DE UMANGO
All these intercalating basic layers are, as stated above, very strongly metamorphosed (corresponding to the „metabasites“ of Hackman). The metamorphism was either of purely plutonic cha- racter, or a result of dynamic influence during later diastrophic periods. The original petrographical character is therefore very diffi- cult or impossible to approve, regarding the circumstance too, that
v i x
d 3
chemical analyses are still lacking. It seems probable, that the
original geological appearance of these rocks are for the most part basic intrusions or sills of diabasic or gabbroid kind, from a time anterior to the orogenetic movements and granite injections.
In order to carry, on a deductive petrographical study of these rocks it is necessary to go to the neighbouring regions, where there are better conditions for such a study. Observations in this respect are, however, still lacking. The occurrence of cambrosilurian fossili- ferous slates, slightly metamorphosed, in the Nevado de Famatina seems to offer a favourable possibility to discover tracks of the primary structures in the basic sills.
The secondary structure of the „metabasites“ of the Umango area is partly of granoblastic habit, due to a complete recrystallisation under abyssal conditions, partly of a crystalline schistosity, caused by a stress influence under contactmetamorphic conditions. In some parts are found also purely kataclastic types.
The amphibolites have undergone a complete recrystallisation and developed a granoblastic structure, especially in the vicinity of peg- matitic veins of larger size, where pneumatolytic action has favoured the former. The types may be described going from east to west.
The above named amphibolite layers in the Cerro Asperecito, situated in the bottom of the Vinchina valley, near to Villa Castelli, are intimately penetrated with small granitic veins and also larger apophyses, and therefore they ought to be completely recrystallized, but a later pression (apparently tertiary movements) has caused a kataclasis. The principal minerals are amphibole and plagioclase, the first named generally developed in individuals of greater size than
ba vd x i | SA Y
the plagioclase. The amphibole shows in the prismatic zone a maxi-
CRYSTALLINE BASEMENT 25
mal extinction angle of nearly 20°, the normal one of ordinary amphi- boles. The pleochroisme varies between grassy-green and greenish- yellow. The prismatic cleavage is well developed. The plagiocla- se is highly granulated by the dynamic influences, forming relatively
small allotriomorphic grains. Albitic twinning lamellae are not well _ developed, so that measures of the extinction angles were difficult _ to realize. Biotite is a more accessory constituent. It is, hovewer,
_ visible in handspecimen. Quartz is very scarce. The rock structure as a whole is distinctly schistose, because the amphibole individuals have been broken up in elongated fragments, following the
prismatic cleavage. These fragments have also been arranged in _ subparallel lines. It may be mentioned, that the c-axis is parallel _ oriented to the general schistosity planes and also the biotite flags. The plagioclase grains are independent of the schistosity and form
a finegrained matrix between the elongated constituents. The crystalline hills of Las Ramaditas on the eastern side of the
Cerro Villa Union show, that the amphibolites play here a very im-
portant part, forming whole hills and ridges. The dark masses of
this rock are distinctly crossed in various directions by granitic and
_ pegmatitic veins of red colour, representing a typical apophyse zone of the Famatina batholith (see later on). The amphibolite is here always highly recrystallized with a well developed parallel struc- ture. The sills lie concordant between quartzitic and thinner li-
mestone layers. Near to the west from the Vinchina valley too, on the eastern slope of the Cerro Potrero Viejo,! where banded rock complexes predomi-
wate, the amphibolite layers are very abundant, playing an important
part in the mentioned banded structure. The amphibolite is of a thinner appearance, because of the concordant penetration with granite
veins. The rock is not only recrystallized, but also partly absorbed _ (lateral banding by assimilation, a phenomenon described later on).
1 Also named Cerro la Pampa.
26 SIERRA DE UMANGO
A microscopical slide of the not assimilated rock shows, that the 4 principal minerals are plagioclase and amphibole. The first named ~ is very altered by sericitic growth, and it is impossible to determine © the extinction angle on oriented sections. The amphibole has a maximal extinction angle of little more than 20° and may con-
sequently be of the ordinary type. The absorption scheme is the
usual too. Small grains of poikilitic quartz occur in the amphibole
individuals. Iron ore is present as small grains thinly scattered ©
through the rock mass. The abyssal body of the southern part of the ridge Sierra de
Umango, called Cerro Cordobés (also Cerro Guecito), contains very - abundantly metamorphosed amphibolites. The rocks occur often — as isolated rootless masses in the granite body, but interstratifica- —
tions between sedimentary schists also exist. A handspecimen of
an amphibolite of the latter kind from the contact with granite shows -
a kind of porphyritic structure, by small spots of a pale mineral
contrasting with the dense dark mass. U. M. the spots or „pheno- crysts“ consist of aggregates of a very dull feldspar. The dark mass is composed chiefly of a green pleochroic amphibole and also of minute feldspar grains. Grains of iron ore are abundantly dispersed
throughout the rock mass. The feldspar aggregates often contain
garnet individuals of reddish colour in their centre, indicating that a recrystallisation of the rock has taken place. An amphibolitic type of a somewhat different composition occurs
oy ee Ka ee ee ees, ee
as an isolated sheet in a „promontory“ or rock cape on the way © between Maz and Villa Union, i. e. on the marked northeastern | corner of the cerro Villa Union. The rock is characterized by a bril- -
liant lustre and an almost black colour on the surface. It forms
a vertical sheet holding red granite on both sides. Already in hand- -
specimen the rock shows a monomineralic composition, an impres- —
sion confirmed by the microscopical investigation. The mineral is | a dark green, intensely lustering one of the amphibole group. Schisto- - sity is not developed in the rock, and the whole structure lacks |
ed ala a ae
CRYSTALLINE BASEMENT ' 27 ör sign of pressure. The amphibole shows the following mi- - eroscopical properties: The colour is very pale greenish — grayish, a somewhat astonishing fact regarding the almost black aspect of the rock surface. The absorption is very slight, and observable only | in orientations parallel to the c- axis. The amphibole-prismatic-cleavage _is well developed. The refringence is high and also the birefringence. The optical character is +. According to the measurements of the angle of extinction, attaining 21°, the amphibole seems to be a variety of pargasite and the whole rock perhaps a product of metaso- _ matic limestone replacement. Quartz is visible only in small quantities. The mighty zone of „schistes lustrées* in the higher part of the “western side of the Sierra de Umango ridge contains, as has been | mentioned, a great number of concordant, regularly alternating sills of an amphibolite schist of some ten meters of thickness. No smaller ramifications or thinner layers occur besides. The appearance is the same as in the Asperecito Hills in the Vinchina valley, but granite veins are not existent here disregarding a broad granite zone on the very crest line of the ridge. _ The rock has a marked schistosity, and the schist planes are covered chiefly with amphibole individuals of an almost black as- pect. On transverse ruptures the rock seems rich in salic minerals, but the whole appears as a kataclastic mass. Microscopically (transverse | cut) one find an extremely crushed structure, formed principally from the breaking off of the feldspar minerals. Also the amphibole has _been dissolved into fragments along the lines of prismatic cleavage and forms subparallel lines of grains. The amphibole has quite _a marked absorption and seems to be of the usual kind. The felds- par grain-mass carries a small amount of quartz too. The size of the grains is variable. The larger ones consist of a plagioclase. The mineralogical composition of the rock seems to correspond to a quartz bearing diorite, once concordantly intruded into the schist mass. Regarding the high amount of later kataclastic metamorphose, see later on (page 99).
Ss) ee > u.
28 SIERRA DE UMANGO
2) Peridotitic rock (sill).
In the eastern part of the group of hills called Las Ramaditas and situated to the east from the Cerro Villa Union, there occurs at the foot of a red sandstone ridge in the crystalline basement a black, coarse grained rock of quite a massive character and running as a sheet concordantly between steeply inclined quartzitic schists.
fue lower serpentine and tert Oxyde Mia Man OLS Ramaditas.
The whole complex is intruded with red granite and pegmatite dikes. In handspecimen the rock is of a coarse granular habit with a black, brilliant mineral as a dominant component. The massive aspect of the rock is remarkable, regarding the circumstance, that the other pregranitic rocks show stress influences.
Microscopically the specimens (see fig. 4) show a peridotitic rock consisting of pyroxene and olivine, both colourless in the slides, but both also partly transformed to green-yellowish ser-
SÅ j ae CRYSTALLINE BASEMENT 29
pentine, The olivine is more intensely altered, often showing ‘only small residuals in a mass of serpentine. Except the last mi- neral, there are as secondary product ferri-oxyde, forming a kind of network in the serpentine mass and doubtless marking the former ‘cleavage of the olivine individuals and also the limits between themselves. The pyroxene has partly a well developed prismatic cleavage. Where this cleavage lacks, it is not always easy to dis- tinguish the two colourless minerals from one another.
| The ordinary amphibolite rocks in the vicinity (Las Ramaditas) are all very schistose. The peridotite lacks all signs of pressure because of the circumstance, that the rock is profoundly altered by serpentinization and because limestone layers in the immediate vicinity have formed a more plastic resistance during the diastrophic pressure.
Other rocks of a similar kind are not found in the Umango „pre- granitic“ series. The described occurence seems therefore to be only a limited one. |
3) Basic, kataclastic schists of doubtful origin.
_ In the Umango area there are in many parts intercalations of „dark schists“ between the sedimentary schists differing from the amphibolitic schists. Megascopically they show kataclastic secondary schistosity, and the original petrographical character is generally very difficult to decipher. All these rocks offer very excellent examples of kataclasis, as will be shown below.
: Many of the types in question have a porphyritic habit. Examining them more carefully, one finds, that this structure is really a por- phyrokataclastic one, the larger minerals are only remnants of lar- ger original grains, while the „groundmass“ is represented by finely crushed rock. The orientation is that, of dynamic flow. In some “cases the schistosity originated at an earlier date than the in- _trusion of the latest (white) granite, and the rock is therefore in “some degree recrystallized.
30 ; SIERRA DE UMANGO
B) THE FAMATINA GRANITE INTRUSION AND ITS EXOGENE PLUTONIC METAMORPHISM.
The Famatina Batholith. | |
The great intrusive body of the Famatina granite, belonsius no
doubt to a very important group of lower palaeozoic granitic unities —
x x
U NONOGASTA
a VILLA UNION + +
N “ Fig. 5. Part of the Famatina batholith, including cambrosilurian schists (black). Shaded: fossilife-
rous parts. The area covers the Nevado. Scale about 1:700,000. Copy of the map of Bodenben- — - der. Later formations not signed. a in the pampean sierras-area, has as most of the other similar bodies ; been laid bare by the erosion to a great extent. This fact is evident 4 by the study of the geological map by Bodenbender (1916) on the Nevado de Famatina, of which here is reproduced a small © copy, showing only the relation between the granite and the in- :
CRYSTALLINE BASEMENT 31
“cluded palaeozoic schists (see fig. 5.) The approximative actual
:
N
i west the body is bordered by the Umango schist area; to the | north it extends far into the mountains surrounding the valley of
3
Pe Te ee ee
extension of the batholith in the surface is, as follows: to the
- Fiambala, in the province of Catamarca, a fact proved by Penck (1915) but shown already on the map by Brackebusch (1891). To the east the granite body reaches probably to the western foot of the Sierra de Velazco, although, according to the map by Bracke- busch, there seem to be outcrops of parts of the gneissic rocks already in the valley hills of Chilecito. According to the map of
- Bodenbender (1901) the granite extends far in the southerly di-
~ rection.
Completely surrounded by this granite lies, as the map (see fig. 5)
_ shows, a complex of metamorphic schists, mica schists, quartz schists
and amphibolitic schists etc. The sedimentary components are in some
\ parts but slightly metamorphosed and do carry fossils here (Kayser _ 1876). The whole complex is therefore considered by Bodenbender (1916) of cambrosilurian age. Besides these fossiliferous parts of _the mass of schists, one generally meets a strong metamorphism,
attaining its maximum on the western foot of the mountain, where mingled and resorbed rocks lie in a mass of pure granite. At the
western wall of the Vinchina valley the plutonic metamorphism is
of a similar intensity as in the Umango schist complex. These
_ phenomena will be treated of in a following chapter.
- Generally the geological appearance of the Famatina intrusive
body has an abyssal habit, exposing irregularily contact lines as a result of a mingling and intense injection of apophyses. Simi-
lar deep seated granitic cores now exposed by the erosion occur
in many parts of the pampean sierras-areas. Already on the map
of Brackebusch (1891) a greater part of these granites are mar-
_ked with ,rocas eruptivas antiguas‘. It cannot yet be stated, if
they are all of nearly the same age, but in many cases a lower
palaeozoic age has been presumed. In the Sierra del Aconquija
32 SIERRA DE UMANGO
Rassmuss (1916) has investigated a granitic body of abyssal appea- — rance, penetrating a complex of precambrian schists, considered by him | to be of a lower palaeozoic age. The shape of the body is probably | that of a batholith with overhead stoping phenomena and a rem-
nant of the roof still visible.
It is necessary to carry on a detailed petrographical investi- gation of the different granitic types of the Famatina batholith and their exogene metamorphic products. The recent communica- tions by Bodenbender (1916) embrace principally stratigraphical observations, and as to myself I had but little occasion to collect specimens here. I have studied in some degree only the western
foot of the Nevado, or the eastern border of the Vinchina valley. — In the following pages I will give a petrographical description of ©
the probably most important types, collected on the western side and also in some parts of the Sierra de Safiogasta. The plutonic metamorphism as concerning the Umango area has of course been studied with greater attention.
a) GRANITE VARIETIES OF THE FAMATINA. During my excursions in the batholith area I got the impression, that the whole body has quite a uniform composition, at least con- cerning the Nevado de Famatina and its southern prolongation, the
Sierra de Safiogasta. The most usual type seems to be a medium- — grained, reddish or grayish biotite granite. Amphibole bearing va-
tities also occur. The structure lacks a parallel orientation, and porphyric development seems to be quite scarce. The reddish
types are often acid or almost aplitic. In the last case a sharper ©
contact with other types may be observed.
The biotite granite, which Bodenbender (1916) mentions from
different localities, seems to be, according to my observations too,
of a wide extension. The colour of the rock varies between reddish —
and gray. A handspecimen of this type from the Quebrada de Tambillos,
CRYSTALLINE BASEMENT 33
on the western slope of the Nevado, shows a medium grained habit, | without traces of kataclasis and of a reddish colour. Mafic minerals are met with to a very small degree. Microscopically the structure seems to be a panallotriomorphic one, with regard to the contours _ between the quartz and feldspar constituents. The feldspar is an or-
thoclase, but generally with a perthitic development showing a kind of undulating extinction. This circumstance indicates, that the perthite
_ probably is of secondary, dynamic origin („dynamic perthite“). Plagio-
clase is found only in a small degree having a zonar structure with transitions between the different zones. Of mafic minerals one finds
only biotite in small spots and also minute grains of iron ore. The rock structure as a whole seems to have quite a slightly altered
consolidation habit, where dynamic influence is shown only by , undulating“ perthite.
Another handspecimen from the Sierra de Safiogasta, on the road from Puerto Alegre to the Safiogasta village, shows a reddish aplitic type with quite large grains of somewhat unequal size. The microscope (see Plate Il, fig. 1) shows a very distinct inhomogeneous structure: small grains mingled with larger ones of allotriomorphic habit. The principal constituents are a reddish, pigmented potash feldspar and quartz. By greater magnification one finds the feldspar
- composed of a cryptoperthite. In some parts of this mineral there are small rounded inclusions of albitic plagioclase, poicilitically pe-
netrating the perthitic structure. Also small quartz grains of similar
- appearance may be observed. These last named ones show a
common optical orientation over greater spots. The structure of the rock has still a fairly primary habit, while only the feldspar seems to have been broken up to some extent, but this might also be a protoclastic phenomenon, so also the aspect of the quartz mass, which is clear and fills up all the sinuous spaces of the feldspar individuals.
A second handspecimen from the Sierra de Safiogasta of a red- dish gray biotite granite of coarse, but inhomogeneous grain, con-
34 SIERRA DE UMANGO
tains biotite megascopically visible in small torn spots, irregularily scattered through the rock mass. Microscopically (see Plate II, fig. 2) there are proois of a strong kataclasis, with the most intense — granulation of the quartz component, the last as a kind of cement 3 filling up the spaces between larger fragments of feldspar. This q latter mineral is viz. lesser crushed than the quartz, an usual phe- 4 nomenon of the kataclasis processes of quartz-feldspar rocks. The 4 dynamic influence is also shown by the development of an undu- lating extinction of the feldspar, frequently in association with a „undulating perthite“, especially in the more elongated corners of i the mineral. Generally it seems to be a plagioclase, although 2 a closer determination is impossible. Mafic minerals are present to —
a small extent, essentially minute biotite spots. This extremely crushed granite seems to be very common in the
Sierra de Safiogasta, and also in other parts of the granite body ~
according to the observations of Bodenbender (1916). The corres-
ponding dynamic processes probably occurred during the tertiary —
diastrophic periods.
In the above mentioned Quebrada de Tambillos on the western slope of the Nevado I discovered also a finegrained aplitic type. The quartz appears in handspecimen dark and well individualized, and the feldspar has a reddish colour. Mafic minerals seem to be completely absent. No microscopical examination were made of this rock. It develops distinct contact lines to the surrounding types of granite, and may perhaps be of a somewhat younger date.
Regarding the occurrences of porphyric varieties in the batholith ©
Bodenbender (1916) mentions a quartz porphyric type from the highest part of the Nevado. I know also such specimens from my
route over the Pampa and Tocino passes. Their occurrences may —
possibly correspond to a higher crustal horizon, because there are -
in the vicinity fossil bearing rocks of the intruded schist mass. A granite type with great feldspar phenocrysts (granite-porphyry) extends over some wider areas on the eastern part of the intrusive
RTL aa ee
ws äv £ CASA ON ete FARA ©
CRYSTALLINE BASEMENT 30
body, outcropping in a series of hilly ridges between Paiman and
Chilecito (see the map by Bodenbender, 1916). ' This type may merely be regarded as an abyssal facies, where the consolidation proceeded in some other way.! The groundmass of the granite-por- phyry is viz. of a coarse grain (normal granitic structure). The pheno- crysts, measuring some cm in length, are orthoclase. They show a
tendency to idiomorphic development. The other components are
quartz, oligoclase-andesine, microcline, apatite, zircon, etc. (Boden- bender). There are seldom pegmatitic differentiates. According to Bodenbender this type dominates also in the region of La Agua- dita (between Los Angulos and Famatina). To the west the rock changes evidently into a more normal granite.
As far as I have observed, the batholith generally seems to carry but few xenoliths except the large schist bodies in the upper parts of the Nevado. Only in some localities on the road between Puerto Alegre and Safiogasta I have noted the occurrence of a dark, quartzitic rock forming breccia floating in the homogeneous granite mass. But approaching the western border of the batholith, i. e., the eastern margin of the Sierra de Umango schist-area, there is, especially in the foothills of crystalline rocks, extending along a line between Villa Castelli and Villa Union, a marked zone of mingling and brecciation phenomena treated of in the following chapter.
b) THE ZONE OF ABYSSAL MINGLING, BRECCIATION AND ASSIMILATION ON THE WESTERN BORDER OF THE FAMATINA BATHOLITH.
Phenomena of plutonic metamorphism in a gigantic scale are well exposed in the foothills at the western slope of the Nevado and continue under the bottom sediments of the Vinchina valley until they appear again in the hills, belonging to the Sierra de Umango-mountains. The metamorphic products, occurring on the eastern side of the valley, are, however, of a special character, due
= The phenocrysts may represent the exceeding part of an ,anchieutectic* gra- nitic-magma-mixture.
36 SIERRA DE UMANGO
to the immediate vicinity of the batholith, so that they may more
conveniently be treated apart from all the other metamorphic rocks, —
occurring in the Sierra de Umango area.
The foothills of the Nevado are rugged, completely barren ridges -
of more longitudinal extension. They rise abruptly over the exten- sive mountain gravel plains and are separated from the granitic slope of the Nevado by similar gravels. The best exposures are to be found opposite the village of Villa Castelli, in the Cerro La Vatea (Batea) (not visible on the map, fig. 5).
The Famatina granite has in this locality penetrated into and dissolved a complex principally of mica schist and amphibolite. In accordance with the general northerly orientation of the Umango
schists the inclusions of the Nevado foothills show the same strike,
in spite of the intense mingling with granitic material. Irregula- rities have been developed more in details.
The manner of intrusion depends on the lithological character of the penetrated rocks. The mica schist has generally been intensely exfoliated along the schistosity-jointing-planes and verbally penetrated with granitic „blood“. This process was accompanied by an intri- cate folding on a small scale indicating a plastic stage of the rock mixture, a contortion without fractures. This phenomenon has been called by Sederholm! ,ptygmatic folding“, a very characte- ristic feature in the deeper archaean terrane in Finland and else- where.
The amphibolite has mostly been shattered into great eruptive breccias with a matrix, composed of acid differentiations from the granitic magma, a very common phenomenon in every such case of injection. ?
1 J. J. Sederholm: Ueber ptygmatische Faltungen. Neues Jahrbuch für Mi- neralogie, etc. Beil. Bd. 36. 2 Heft. Sept. 1913. Page 491.
2 Adams and Barlow: The Geology of the Haliburton and Bancroft Areas, Province of Ontario. Canada. Dept of Mines. Geol. Surv. Branch. Mem. 6. 1910.
J. J. Sederholm: Om granit och gneis. Bull. Comm. Géol. de Finlande. N:o 23. Helsingfors 1907. With an english summary.
a a a a | H al oy bet | y Vi kv
| he N a f (
EIN N u =
Pe aes ws
EEE EN
kn ee —
rnit TIERE
FREE res te EVEN
Se Sete
CRYSTALLINE BASEMENT 31
The injection process was not only limited to a mingling of the
a granite magma with the schist mass. There are also on a gigantic
scale the signs of an resorption, in the extreme cases conducing to almost homogeneous rocks of different composition. Natu- tally all stages on the way to a complete resorption may be observed. The mica schist passes into a dark gneiss of quite a homogeneous habit, and the amphibolite breccias are gra- dually dissolved into a chaotic mixture, or ,stew“, of resorbed ot partly resorbed xenoliths and granite. The extremely resorbed amphibolite or, more correctly, a basic secondary granite magma, has, as stated later on, a considerable extension in this zone.
The incompletely absorbed masses, the inhomogeneous mixtures show often a roughly parallel structure; the fragments and the almost resorbed spots are oriented in common direction, indicating, that the whole mass was in a flowing motion before consolida- tion. Also the completely absorbed masses develop in some cases flow structure.
Petrographical characteristics.
The petrographical description of this zone may begin with the amphibolite and its metamorphic products, because they play the most important part in the general composition of the zone. |
The great amphibolite breccias show often the unchanged amphi bolite without other signs of the granite action, but a complete recrystallisation. The fragments are angular and often of a consi- derable size. The contours are sharp against the matrix, which is a pegmatitic, pale coloured granite, carrying isolated amphibole and epidote of greater size. The structure of the rock fragments is granoblastic and often coarse, and towards the edge the grains increase in size, probably by influence of the pegmatite matrix. The borders are therefore not seldom on the rock surfaces marked with a black fringe (great amphibole individuals).
38 SIERRA DE UMANGO
The principal minerals, are feldspar, very altered, probably always plagioclase, and amphibole of ordinary habit. The plagio- —
clase forms partly porphyroblastic phenocrysts, with extinction angle on cuts perpendicular to M and P corresponding to Ab,,An,). The
contours of the mineral components are generally sinuous. No tracks ~
of pressure are visible, and it is possible, that the rock after some earlier crushing process has undergone a complete recrystallisation with a hornfels structure. Iron ores in minute grains are scattered over the whole rock mass. Also biotite is present, showing strong absorption and an intimate association with the amphibole. It is
highly probable, that the biotite is only a metasomatic product of |
the amphibole, originated by influence of the granitic magma. This phenomenon will be treated of at an another occasion later on. All stages of transformation of amphibole to biotite are represented,
and the extremely dissolved breccias show fragments wholly com-.
posed of biotite rock.
The last product of the assimilation process of the amphibolite by granite is quite a homogeneous rock of true magmatic aspect, a syntectic diorite. This rock was already observed by Bodenben- der and sometimes called by him ,tonalite‘.? It is surprising wide spread in the present rock surface of the foothills. It forms the principal rock of the zone.
This remarkable rock is gray in colour, of large grain and quite rough, and its surface weathers in a peculiar way, forming small kettles and hollows, like results of a desert deflation. Dark inclu- sions of mafic mineral aggregates are common. By its whole ap- pearance the rock is well distinguishable from the ordinary Fama-
tina granite. Contact lines between these two have not yet been
studied.
Regarded from a closer distance or as handspecimen the dioritic —
1 See the Plate III, fig. 1.
2 According to Bodenbenders handspecimens from the Nevado, kept in the -
Museum of the Direction of Mines.
EEE ELTERN RT a a BSE TERN rer
en TÖTET SF
ee
I
Ri ace ae att
CRYSTALLINE BASEMENT | 39
rock shows a remarkable inhomogeneity due to the appearance of
“biotite spots of irregular size and distribution. Somewhere the biotite concentrates to greater streaks or Schlieren. These are, as: was stated above, often subparallel, evidently by flow du-
ting the consolidation. Microscopically the rock is quite quartz- bearing. According to measurements of the angle of extinction
i in cuts perpendicular to M and P of the principal mineral, a white
_ plagioclase feldspar, it has the composition Ab,,An,,. The chief
3 mafic component is, as has been stated, a biotite with a strong
BETEN |
absorption. This mineral is frequently penetrated with poicilitic
_ grains of quartz, with common areal optical orientation. The border
zone between the feldspar and the quartz develops a myrmekitic
intergrowth. The structure of the rock is, however, in gene-
tal allotriomorphic. Only small scattered apatite individuals show
: idiomorphic forms. The quartz is of jagged contours, probably
due to recrystallisation, while traces of pressure are not visible.
- Amphibole occurs to a reduced quantity as insignificant individuals.
The biotite is partly transformed to an opaque mass. Iron ore
also occurs in minute spots.
Aside from the basic concentrations (biotite bearing masses) in
this rock there are numerous acid Schlieren, apparently real diffe- rentiation products of the syntectic magma. These acid varieties are of a pale red colour and have no precise limits towards the main
rock. The shape of the Schlieren is irregular and may be kept
apart from the crosscutting pegmatite dikes of the last eruption
phase of the batholith, described below.
To the west, towards the Umango schist body, the syntectic,
| quartzbearing diorite shows a marked irruptive contact, well visible
in the Asperecito hills to the west of Villa Castelli.
A dark gneissic rock, occurring as quite a long zone in the abrupt western slope of the Cerro La Vatea, is very probably of syntectic origin too. In handspecimen the rock seems to be more a basic granite with a marked parallel structure, visible on the rock-
40 SIERRA DE UMANGO
surface. The fact, that the rock in question appears in localities, where the mica schist is most intensely penetrated with granite, 3 makes it probable, that it is a product of complete absorption of the named schist. — The microscope reveals the following mi- a
neral composition: the feldspar is partly a microcline, partly a plagioclase of ordinary, dioritic habit. A biotite with strong pleo- chroism is abundant and partly transformed to a fibrous substance, showing a high refringence and birefringence. Its colour is trans- parent, light pale brownish. The mineral is evidently the sillima- nite-variety called fibrolite. Also muscovite is visible, almost in- variably associated with biotite. — The structure of the rock is
generally evengrained without signs of kataclasis. Only the fibrolite I
has partly a dynamic flow arrangement as shown on microphoto- graph (Plate III, fig. 2). The transformation of biotite to fibrolite is very probably a dynamic phenomenon too.!
c) CROSSCUTTING PEGMATITIC DIKES. The Famatina last injection phase.
The sharply crosscutting, broad pegmatitic dikes here are younger than all the mingled and molten rocks in the abyssal zone on the eastern side of the Vinchina valley. Doubtless they due their origin, however, to the Famatina injection process and form its last „acid“ phase, because of the fact, that the dikes show transitions to the ordinary aplitic granite. The crosscutting character and the conspi- cuously flat lying position of them indicate, however, that there must be a hiatus between the principal intrusion act and the erup- tion of these apophyses; in other words, that there was already a
1 The principal occurrence of sillimanite is, as known, in exogene contact rocks
and also as inclusions in a granitic magma (Cfr. Adams and Barlow, l.c. page —
131 a. 0.). The typical development of fibrolite is, according to Rosenbusch (Mikr. Phys. der Min. u. Gest. Teil II. Stuttgart 1905. Page 137), in the form of a dynamic transformation product of biotite. In the present case no simple contact influence has produced the alteration because the very rock is an syntectic one. The dynamic process is probably of tertiary age.
u er ae
ER
ae Ta PTE NEC) a eat BEE en
aa rest ee A i ig nn
te
DE
u
ee ae
ac A
a é ig +
CRYSTALLINE BASEMENT 41
stage of beginning consolidation in progress, when the pegmatitic pe cut through the zone in question.
"> These dikes, broad and light coloured, very markedly visible in the | rock ground, are highly differentiated. The border zones are formed : of a red feldspar and the midst of the dike is filled with milkwhite quartz. Frequently there is also a smaller exterior border outside h the feldspar mass consisting of an evengrained aplite. The limits between the feldspar and the quartz masses are quite distinct. The i thickness of the dikes varies ; much and attains several times some meters. It may be noted, that the differentiation appears with the same distinction in dikes of various breadth.
The quartz mass contains
often well developed crystals
of a black tourmaline (schorl). The feldspar is a red ortho- clase and forms, when isola-
ted, great individuals (see
Fig. 6. Great individuals of red potash feldspar R 3 é ying in a milkwhite quartz mass. The cleavage lig. 6), never mingled with ee nel Guar, |The habit of these Om Wilke: Ural individuals is idiomorphic in some degree, showing partly the pinacoids well developed. Pneu- ‘matolytic minerals are scarcely existent, except the above named tourmaline. Beryl was once found in a pegmatite and was of greater dimensions with the prismatic faces striated (Cerro La Vatea). The flat lying position of the dikes is difficult to explain with regard to the general steep inclination of the schists intruded with | granite in this zone. The pegmatitic eruptive material has appa- rently followed independent fissures, opened subsequently to the
consolidation of the mingled-rock-zone.
The origin of pegmatites has been regarded, as known, partly as
42 SIERRA DE UMANGO
plutonic, partly as ,neptunic".t All transitions between these two q
have been; found as well. Keyes? in his paper on the Mary-
land acid eruptive rocks distinguishes between intrusion- and 4 separation-pegmatites. A similar division was also pointed by q Holmquist.® Regarding the Famatina granite-pegmatites above ~
described, they are all of a purely plutonic character, because
they always form typical apophyses, crosscutting or following the 4
schistosity planes of the older rocks. The circumstance, that the pegmatite dikes near the batholith show an extreme differentiation
of the quartz and feldspar components, may be explained by 4
assuming the assistance of katalytical gases, causing greater molecu- lar mobility at the beginning of the consolidation.
d) PLUTONIC METAMORPHISM IN THE SIERRA DE UMANGO SCHIST-BODY.
The western sector of the Famatina batholith injection aureole. It has been stated above, that the Famatina batholith is to a great extent exposed in the surface by a deeply cutting erosion. The consequence of this fact is, that also the cut through the aureole must be of a considerable breadth. A part of this aureole forms
to the west the Sierra de Umango schist- area with its innumerable.
granitic and pegmatitic dikes. In peripherical direction (to the west) it cannot be stated any decreasing intensity of the exogene con- tact influences, because the aureole is on this side cut off by a great fault. Also later intrusions have altered the succession of the contact zones referring to the batholith. Almost a similar intensity of plutonic metamorphism dominates thence the whole area, except
1 C.R van Hise: A treatise on Metamorphism. U. S. Geol. Survey. Mono- graph 47. Wash. 1904. Page 720 and foll.
2 C. R. Keyes: The origin and relations of the Central Maryland granites. U.S Geol. Survey. Ann. Rep. XV. Wash. 1893—94.
3 P. J. Holmquist: The Archaean Geology of the Coast-regions of Stockholm. Geol. Foren. i Stockholm Förhandl. Band 32. Haft. 4. 1910. (Also Guide N:o 15 of the Int. Geol. Congress, Stockholm 1910.) Page 801 and foll.
POE eT eRe CE een YES
CRYSTALLINE BASEMENT 43
some apparently downfaulted schist-zones, where granite is lacking. These zones belong probably to a higher crustal horizon.
The degree of metamorphism in the Umango area never reaches _the intensity of the abyssal phenomena as in the above described zone on the very border of the batholith, although smaller assimilations also occur here and there.
i In the following pages only the products of the Famatina injec- tion will be treated of, while all the cases of alterations caused by
a younger granite-syenite will be described afterwards.
| The Famatina granite appears in the Umango schist body prin- _ cipally as crosscutting or concordant dikes of red granite aplite or pegmatite without differentiation. In some parts we find also broader zones of aplitic granite between the schists, showing at its borders
EEE
small assimilation phenomena.
The eastern mountain slopes of the Mo area form a zone of some different character, compared with the zone to the east, at the foot of the Nevado. The schists are in the above named slopes 4 not yet dissolved from their mutual connection, and the strike is in detail but slightly disturbed. Melting phenomena are of quite a limited extent. The granite appears here as crosscutting dikes,
eva or
=...
oe? FA
4 running in several directions. Only to a smaller extent a parallel injection may be observed. The dikes are of two petrographical _ kinds: red, evengrained aplite and red pegmatite, the later of an eutectic structure. These are more important and mightier than the _ aplite apophyses.
_ The pegmatites are of a crom red colour and show no complete _ differentiation in the quartz- feldspar components. The fact that these _ pegmatites are „graphically“ developed, indicates some different conditions of consolidation than those in the zone nearest to the batholith, where katalytic gases and a slower cooling formed
favourable conditions to the development of a zonar arrangement
rae NN
_ of the components. _ The feldspar in the graphic pegmatite is of backstone red colour
ae ee ast © 5
44 SIERRA DE UMANGO
and has brilliant lustering cleavage surfaces, where one may observe the presence of small facets. Microscopically are seen, that these facets find their explanation in the structure of the mineral itself. j The principal part of the feldspar is orthoclase in cryptoperthitic in- 2 tergrowth with albite. In this mass there are zones of microcline q with transition to the cryptoperthite. These microcline zones seem q to be the result of a pression, identical with the „undulating extinc- 4 tion" in the pressed quartz. This kind of secondary microcline 4 might be called „dynamic microcline“.! The quartz component of the pegmatites is always very clear and shows rounded contours ©
Må
against the feldspar. In spite of the signs of dynamic affect in the °
3 feldspar no pression phenomena are observed in the quartz. The red granite apophyses, consisting of an evengrained aplite, are, as said, of a more reduced breadth. They are contemporaneous with the pegmatite dikes and show in parts transitions to them. The principal occurrences of the aplite dikes are in the mentioned zone i of the eastern slopes of the Umango mountains, especially in the j hills of Las Ramaditas. The structure of the aplite is but slightly | crushed. Where the dikes cut across a limestone layer, a small | contact zone of „skarn“, i. e., garnet and epidote, may be observed. ; The intruded quartzitic layers contain red garnet in smaller indi- viduals. In the crystalline rock-hills of Asperecito (to the north) are found pegmatitic dikes of considerable size, and also aplite. The pegmatite is very rich in muscovite, partly crystallized in sphaeri- cal aggregates, partly in the form of greater sheets, exploited in — past times. Turmaline (schorl) is also present here. Magnetite — occurs in smaller well developed individuals. |
To the west, up to the higher mountains, the pegmatite-
1 About the formation of secondary microcline see: A. Harker: Natural History of Igneous Rocks. London 1909. Pages 259—260, and J. H. L. Vogt in Tschermaks Min. u. Petrogr. Mitt. (2). Vol. XXIV. Pages 537—-541; an interesting application is to be found also in the paper of H. L. Alling: The mineralogy of the feldspars. Journ. of Geol. 29. 1921. p. 275.
CRYSTALLINE BASEMENT 45
aplite dikes, especially the crosscutting ones are getting more , scarce, and in some schist-zones red granite is completely lacking. But in some other parts there are, however, not dikes, but _ whole zones of red granite. Also real bodies of granite of this . appear, giving an abyssal character to the locality. Such Strongly sranitized areas to the west are: Cerro Cordobés, Cerro la Pampa (northern part of the Cerro Potrero Viejo) and the western slope of the Cerro Cacho, down to the tertiary fault line . the Cordilleran border. In the last; named area there are also _ porphyric varieties, but the dominant rock is a red granite. = The structures of all these granite bodies are not described here,
because they are almost crushed by later movements, and these phenomena will be treated of later on (pag. 50 and foll.).
- ©) AN EARLY PALAEOZOIC REGIONAL METAMORPHISM, SUBSEQUENT TO THE FAMATINA INTRUSION.
- In the Umango area there are signs of orogenetic disturbances , accompanied by an intrusion of a white granite-pegmatite and syenite of apparently younger date, than the Famatina intrusion. The above “mentioned white eruptives cut across the red granite in marked dikes, although the most common appearance of the former is a parallel banding in the schists to the west, apart from the Famatina batholith with its network of apophyses. Simultaneously with this younger intrusion a lateral pressure (stress) occurred, causing folds in the limestone banks and also a marked banding over broader ‘zones between amphibolite and other schists and concordant dikes of white granite (syenite). This injection has as a whole a less d markedly abyssal character than the Famatina plutonic phase and belongs probably to a somewhat higher crustal horizon, than the - last mentioned. It is evident that the white granite-syenite intrusion occurred in prepermian („pre-Gondwana“) time, because of the fact,
u ee Zee RR
FINS u ee eh
46 . SIERRA DE UMANGO
-
that the lower permian tillites rest directly over this twice granitized | | ut
basement.
Ch
As was already stated in short, the white granite-syenite intrusions fj 0!”
are met with principally in the western parts of the Umango area,
outcropping as broader zones in the Sierra de Umango ridge and 9"
the Cerro Potrero Viejo. The whole intervening space between the former ridge and the Cerro Cacho is also rich of such dikes. In the last mentioned mountain there are similar dikes too, but a red
granite, however, dominates. Only in one ‘part the intrusion of Au
stance makes it evident, that the intrusion was contemporaneous
with the tangential pressure.
The folding and the banding are also phenomena characterizing | hi
the appearance of the white granite-syenite. To the east, in the | pal
region of the Famatina batholith there are no signs of lateral stress, conc
and a younger granite seems to be lacking there too. Boden- | bender mentions no occurrence of this intrusive from there, and § as to myself, I have never discovered any dikes of this rock inside | Fin di
the batholith.
Nothing further are known regarding the occurrences of youn- | ger (post-Famatina) granites in the pampean sierras-area, be- 7 cause the study of the chronology of the old intrusions in this area :
is still a task never touched upon. According to a verbal com-
munication by Rassmuss, he has seen traces of a younger intru- — sion in the Sierra de Velazco, a white granite crossing the red one, the ~
latter corresponding to the Famatina granite.
a) MANNER OF INTRUSION OF YOUNGER GRANITES AND SYENITES.
As already stated, the younger intrusives show a less conspicuous | N: abyssal character than the Famatina red granite, and they depend | to a considerable degree in their manner of intrusion on the struc- —
CRYSTALLINE BASEMENT 47
tural planes of the schist mass. Crosscutting dikes are occassionally seen, but more irregular bodies with sinuous contacts and numerous -apophyses are not met with. Brecciation, melting and assimilation hs are also odd phenomena, except a certain class of banding, Fk where a lateral assimilation on a very reduced scale has suc- j ceeded. The whole appearance of the intrusion depended appa- — tently upon a lateral controlling pressure, causing the mentioned i parallel injections along the planes of weakness. The areal exten- sion of the resulting banded rocks is a very considerable one. Alt- hough they are of the same origin as many common parallel ar- : terites in archaean terranes, the banding in the Umango area shows a somewhat different character, due to an already mentioned lateral a assimilation along the granite veins. The lateral assimilation is com- mon especially where the granite intrudes amphibolite schists, from which results a product of intermediary composition. This last is _ also banded, evidently a result of parallel movements under stress conditions. Also at the borders of other schists such a banding _ was produced. The banding phenomenon has been studied already long time ago _ in different archaean terranes. The following two classes may be 4 distinguished: 1) Banding produced by flowing in a primary magma _ with the formation of „Schlieren“ or streaks of different composition. : The well known banded gabbro of Skye, Scotland, described by Geikie and Teall! belongs to this type. 2) Banding produced by _ the injection of magma along the structural planes of a schist mass. _ This type is the more important one and is of regional extension, as is shown by investigations in many parts of the world.”
1A. Geikie and J. J. H. Teall: Banded Igneous Rocks of Skye. Quarterly _ Journal of the Geological Society of London. Vol. L. 1894. | 2J. J. Sederholm: Om granit och gneis. Bull. Comm. Géol. de Finlande. | N:o 23. Helsingfors 1907.
P. J. Holmquist: Zur Petrographie und Geologie von Ornö Hufvud. Bull. Geol. Inst. of Upsala. Vol. X. 1910.
Adams and Barlow: Geol. of the Haliburton and Bancroft Areas, pro-
48 | ~SIERRA DE UMANGO
The banded rocks in the Sierra de Umango area may be regarded rather as a combination of the two classes characterized above: flow-banding and parallel injection. In the cases, when the lateral : assimilation has progressed very far, i. e, when the invaded ~ rock to a greater extent has been dissolved in the granite, one ob- | serves the development of a rock banded by flow too.
I have studied the case of lateral assimilation of amphibolite- schists by granite. The whole injection-banding and assimilation is well exposed in the Quebrada Agua de los Caballos, on the eastern slope of the Cerro Potrero Viejo. Here may be observed,
Pre SN pd et ee EA eS AR
that the amphibolite becomes by and by more acid by lateral mingling, not continuously, however, but in the form of bands of decreasing basicity. Simultaneously the amphibole changes into
ee a a FE u u
biotite (Plate IV, fig. 1). The last assimilation product is a biotite gneiss. |
About the transformation of amphibole to biotite Van Hise! says: ,The change from hornblende to biotite is a much deeper seated alteration (compared with the alteration to chlorite), parti- cularly in connection with profound mechanical action“. Regar- | ding the change caused by granite assimilation (addition of salic — material) the same author states: „the feldspar........ frequently | furnishes the potassia, part of the aluminia and silica, while the amphiboles frequently furnish a part of the magnesia, aluminia and silica“ to form the biotite. The conditions in the Sierra de - Umango may offer a combination of the two possibilities, above cited.
vince of Ontario, Canada. Dept. of Mines. Geol. Survey. Mem. o. Ottawa, 1910. P. 73.
R. A. Daly: Geol. Reconnaissance between Golden and Kamloops, B. C., along the Canadian Pacific Railway. Ibidem, N:o 68. Ottawa 1915. Pages 34 and toll.
C. N. Fenner: The mode of formation of certain gneisses in the highlands of — New Jersey. Journal of Geol. Vol. XXII. Chicago 1914. A. o.
1 C.R. Van Hise: A Treatise on Metamorphism. Page 290.
eS es
Ss a Zu ra Te un ln En Bussen ae Zi
CRYSTALLINE BASEMENT 49
b) PETROGRAPHICAL CHARACTERISTICS OF THE YOUNGER INTRUSIVES.
The greater bodies of this class of intrusive rocks in the Sierra de Umango area seem to consist of a white-gray syenite, quite large grained and very poor in mafic minerals. Only where there are inclusions of some basic, strange rocks, an assimilation product
of a more basic character may be observed. Where these intrusives
appear in form of dikes, they: are mostly granites and pegma- tites, all of acid character. Where crosscutting dikes occur, they consist of a finegrained aplite.
The easternmost occurrences of the white intrusives are met with in the crystalline hills of Las Ramaditas, where crosscutting dikes of the aplite and also of white pegmatite are visible on quite a large scale. The pegmatites are rich in quartz, and have been but slightly
‚crushed by later movements, thus contrasting with the occurrences
of the same rock further to the west (except where the rock lies
between limestone banks). The aplitic dikes are of quite fine even
grain and show microscopically a somewhat crushed structure, pro- bably depending upon a protoclasis, owing to the circumstance, that the dikes are cutting straight through the schist mass, without flex- ures or ruptures.
Further to the west, in the abrupt eastern slope of the Cerro Villa Union, opposite the hills of Las Ramaditas, there occurs a boss of a white syenite. I had no opportunity to study this intrusive body and its contact lines. I only collected some speci- mens from a river bed leading down the slope from the heart of this massive. The rock consists, according to these boulders (huge masses), of an almost monomineralic kind of evengrained intrusive, an albite-oligoclase mass. The feldspars are generally very altered by sericitization. Muscovite is also a constituent, but not abundant. Quartz was not detected, neither in handspecimen nor in mic- roscopical slides. The structure seems to be completely free from crushing phenomena.
50 SIERRA DE UMANGO
The structures of the white intrusives, occurring more to the west - (in the high ridge of the Sierra de Umango etc.), will be treated of -
later on, as they all belong to secondary structures of Öyramira origin, caused by the tertiary movements.
D) KATACLASTIC STRUCTURES OF THE BASEMENT ROCKS, PRODUCED PRINCIPALLY BY LATER (TERTIARY) OROGENETIC MOVEMENTS.
As will be shown! later on, the whole Sierra de Umango area |
was affected during tertiary times by orogenetic movements. In an
earlier phase these manifested themselves by a tangential pressure
from the} west, causing am. o. the folding-tectonic in „the inter- montane zone“. It will also be shown, that the Umango area forms a part of „Foreland“ or „avant pays“ to the andean chain.
These eastward pushing tangential movements produced not only
different disturbances in the continental sandstone cover. The —
crystalline basement was also strongly affected and a regional kata- clasis developed in it. The following petrographical description
may be a study of the crushing phenomena of the rock minerals,
referring to the well known researches of Leith.”
It might be supposed, that some of the below described secondary |
structures may originate from an earlier diastrophic period than
the tertiary, for instance, from the same phase, when the white
granite was intruded. It seems, however, probable, that the kata- |
clasis developed during this period, was brought to recrystallisa- tion by the subsequent intrusion of the white granite.
The tertiary tangential pressure released along NNE running over- thrust planes and broke up the Sierra de Umango crystalline base-
1 Tertiary tectonics, in Part II.
1905.
2C.K. Leith, Rock cleavage. U. S. Geol. Survey. Bulletin 239. Washington
Pg nat ee ay ee, ee
PE ee ea Se OES er ee ie
of: „ -
CRYSTALLINE BASEMENT 51
ment into a number of great fault blocks. The downpressed sandstone- remnants lie between the blocks in the throws. The crushing of the basement rocks apparently went on through the whole complex. The intensity varies, however, considerably. The most complete granulation occurred along the overthrust planes, in whose im- mediate vicinity a mylonitic breccia was developed. The crushing and tearing was intense also where schistose rocks dominate, while in the granite stocks conspicuous signs of pressure have been observed more locally. Generally, the degree of deformation may be said to decrease towards the east. In the batholith of the Nevado the granulation of the granite structures seems to have been of a more local nature, while the bulk of the eruptives shows quite a well preserved consolidation habit.
a) THE CRUSHED, GRANULATED GRANITES.
In the foregoing pages the granites of the Sierra de Umango area are petrographically characterized only to a small extent, because primary granitic structures seldom occur. The greater part of the granite outcrops shows more or less conspicuous signs of dynami- cal deformation.
In the following pages may be examined the secondary structures of the granites in the order as they outcrop, going from east to west.
Above have been treated of the Famatina batholith types. They show generally quite a primary habit. Passing the valley to the Sierra de Umango side, in the next zone (schists with crossing apophyses) conspicuous crushing features are lacking (graphic pegmatites!)
‚But some miles farther west, in the eastern slope of the Cerro
Villa Union, there are already strong pressure phenomena. The concordant granite veins in the schists have been torn off into lenses and the structure is completely granulated. It is here seen a typically developed banded zone. As a good example of a crushed concordant pegmatite vein an occurrence from the western foot of the mentioned mountain (Carrizal) may be described. Already in
52 SIERRA DE UMANGO
handspecimen the rock shows a marked dynamic flow structure,
but microscopically it is the most typical of this kind: the parallel is produced by the alternate of strokes of quartz grains and feld-
spar + muscovite grains, the whole showing a great similarity of
the parallel, crushed granite structures described by Leith.! The strokes have been formed by the breaking off of larger grains. The quartz bands have afterwards undergone a recrystallisation in some
degree (see microphotograph, Plate V, fig. 1). The granite described -
lies in a mica schist complex, strongly pressed too. — In the whole mass of the Cerro Villa Union the pressure phenomena are quite of the same intensity.
On the crest of the Sierra de Umango ridge there is, as has been mentioned, a zone of red, aplitic granite, concordantly penetrating the schists. In handspecimen the granite has a marked parallel structure („gneiss-granite“) apparently of dynamical character. Micro- scopically the rock shows a middle stage of granulation with many variations in size of the grains. The quartz has afterwards recrystallized to some extent, as seen from the sinuous contours of some larger, unbroken grains. The feldspar is partly albite-twinned and very altered by chemical decomposition, partly a badiy deve- loped microcline. Mafic minerals are lacking with exception of small spots of biotite.
At the western foot of the same mountain ridge, near to the Ar- royo Gaucho, following his border, there is outcropping quite a large zone of a red, aplitic granite, bordered upward by the luste- ring mica-schist-zone. Because the granite zone runs almost confor- mably with a tertiary dislocation line (the Pirca dislocation), this rock has undergone an intense crushing. The kataclasis seems to increase downward in the mountain slope, being most intense along the very fault line (at the foot). The extremely crushed specimen shows a habit of an „aplitic schist“ (Plate V, fig. 2). A transversal
1 Rock cleavage. U. S. Geol. Survey. Bull. 239. 1905.
f i #
I an en REN
i
2
2:
a
RS if
BR
RR EP U > Fa a
ae ep ng ee
ee te ee RO
t
i ae A då
eT
ee ee
Y
IR ER
Euer;
SEE NONE DEE DONE EE
CRYSTALLINE BASEMENT 53
microscopical slide of this type demonstrates a very fine granulation of the quartz and feldspar components. The secondary grains are all isometric in shape and the parallel structure is less conspicuous
than in handspecimens. Flat lenses of a secondary, opaque substance can be seen, probably the last remnant of some mafic minerals (amphibole?).
In the Cerro Cacho complex all the granitic stocks of longitudinal shape and the concordant dikes have been affected by the tertiary and partly also by some older movements. The most intense cru- shing appears near to the western border of the complex, where the bordering fault-line runs (to the west from Quebrada Cacho and Quebrada Agua de la Piedra). On the eastern slope of the moun- tain, where the seleniure mining area is situated,’ there is a large, longitudinal stock of aplitic granite, penetrating amphibolite schists and limestone banks. The granite has been transformed into gneiss by pression phenomena. That the pression here really belongs to an older phase is seen from the circumstance, that the granite has been folded simultaneously with the surrounding schists, and is penetrated by a dike of white pegmatite. A handspecimen of the red granite shows a distinct parallel structure, but not a schistose one. The mineralogical composition of the rock seems to be that of an eutectical mixture. The graphic texture has indeed been torn off. But in spite of the strong pressure the quartz and feldspar components have not been mingled together to a higher degree, a fact, proved also microscopically. The microscopical examination shows, that the resulted grain of the rock is a ,double“ one, that is, every individual has been broken off into smaller fragments holding to some extent their original position. This is the case especially with the feldspar grains, while the quartz mostly forms a cementering or filling mass between the feldspar aggregates. This quartz mass has a dynamical flow arrangement, but generally with an oblique di-
1 See the concluding chapter, Part II.
54 SIERRA DE UMANGO
rection compared with the general flow direction of the rock. The feldspar grains are very altered, and belong probably all to the potash feldspar group. In an extreme case of pressure the aplite granite, or more exactly the graphic granite, has been torn off to a kind of a fine banded rock, like the banded hälleflinta, consisting of alternating layers of quartz and feldspar. The structural direction is moreover marked by the appearance of parallel biotite flags.
b) DEFORMATIONS IN THE SEDIMENTARY SCHISTS. Among the sedimentary schists of the Sierra de Umango base- ment the mica schists are the most affected by the tectonical mo- vements. The general habit of these rocks indicates with complete
evidence, that they must have undergone a strong kataclastic meta-
morphism. The development of quartz lenses, folds and faults is found everywhere, and also microscopically the helicitic structure. Especially the abundant manifestations of secondary quartz-lenses
and layers in all parts of the area are a characteristical feature of
the powerful dynamical influence in these kinds of rocks.
A specimen of the lustering mica schist from the western side of the high ridge of the Sierra de Umango demonstrates microsco- pically a strong folding of the mica planes, conducing in extreme cases to the „Slip cleavage“ (Leith). The rock mass between the mica planes is composed of small quartz grains.
The forming of quartz lenses between the schist planes conduces to a lenticular structure, a characteristical phenomenon in the zone of katamorphism. This phenomenon was evidently a tertiary one and not contemporaneous with the granite injections, as the lenses never show connection with acid granitic apophyses. Sometimes these secondary quartz lenses are thinned out so as to form regular layers alternating with the mica planes, thus producing a type si- milar to the „Lagenglimmerschiefer“ of Rosenbusch.
The circumstance, that the mica schist-zones are especially affected by the gliding and crushing phenomena, may be explained by assu-
ya er EAS
CRYSTALLINE BASEMENT ; 55
ming the named zones as ways of weakness, the mica planes having _ played the part of gliding surfaces.
On a very much smaller scale we find the pression phenomena i
_ developed in the quartzitic schists, where generally every kind of _ crushing is lacking. Also the grains of quartz are individually well conserved (after the last recrystallization) and no signs of ,undu- lating extinction’ can be detected. The slight mechanical defor-
ua 6 Ene ro
_ mation can be seen already in handspecimen, whose surface shows a distinct glassy apparition (in the transverse cuts). This remarkable - consistense of the rock can be explained in general by a greater -hardness and coherence characterizing this kind of crystalline schists. The crystalline limestone, occurring, as mentioned, principally as folded banks in the western part of the Umango area, was intruded by white granite and pegmatite simultaneous with the folding process.
By this process the rock became completely recrystallized, and afterward — during the tertiary time — no crushing deformations
have occurred, except in some isolated localities, where faults cut across the banks.
c) KATACLASIS IN SOME „DARK SCHISTS“.
A group of dark schistose rocks, characterized above petrographi- cally, affords a very excellent example of the tertiary dynamic meta- morphism. They are sometimes of a doubtiul origin. Compared with the ordinary amphibolites they are much more granulated, a fact preventing essentially the examination of the real nature of the rocks. Microscopically the dark schists often develop a porphyroclastic structure (Plate VI, fig. 1). The larger grains are composed of plagioclase and amphibole. When quartz occurs, it has been so intensely granulated, that it forms a kind of groundmass. This fine detritic mass has a flow arrangement, bending around the larger grains and containing also abundant minute amphibole and biotite fragments of elongated forms, especially marking the flowage direc- tions. An interesting example is illustrated on plate VI, fig. 2, where
06 SIERRA DE UMANGO
biotite plays the part of „phenocrysts“ and greater lenses, while
the fine granulated mesostasis is composed of quartz and sericite.!
| |
4
E) GENERALITIES ON THE GEOLOGICAL POSITION OF THE |
SIERRA DE UMANGO AREA.
As a consequence of the scarceness of our knowledge of the ~
geology of the vast area composed by crystalline, partly sedimen- tary and partly igneous, rocks of the central part of the Argentine
Republic, appearing in the so called „Sierras pampeanas“, reaching —
to the very border of the Andes, it will be a difficult task by a limited surveying in this area (the Sierra de Umango) to explain any broader tectonical and structural relations. Working in the Sierra de Umango area the writer did not get any occasion to make excursions into the surrounding countries so as to obtain more general views. The existing litterature and the maps of the neighbouring districts, lately prepared, have, however, facilitated an attempt to a somewhat broader synopsis, although the results may be in many respects only hypothetical ones.
The general orientation of the metamorphic schists in the „Sier-
ras pampeanas“ has been fixed as running nearly N—S over wide ©
areas. Also in regions near to the Andes this crystalline basement shows a similar behaviour. The schists of the Sierra de Famatina, the Sierra de Umango and also of the Sierra Pié de Palo farther south are roughly trended into this general orientation.
Granitic bosses and batholiths of various extension are intruded into the gneiss and schist complexes at various times and have caused a regional metamorphic, and in many cases ultrametamorphic change of the preexistant rocks. Similar kind of abyssal metamor- phism has overgone, as has been shown above, also the Famatina— Umango schists, principally by intrusion of the Famatina batholith.
1 See also Plate VII, figg. 1 and 2.
WEGEN (Be ER EEE ai er a i
SP IA
Sg ao ee
CRYSTALLINE BASEMENT 97
4 The abyssal character of the rocks dominates westward to the very
border of the Andes, where suddenly clastic rocks appear. The limit between the crystalline area and the clastic series of this part of the Andes may thence be represented by a great fault line.
A similar contrast exists between the lithological composition of the Sierra Pié de Palo and the near lying parts of the Precordillera.
The rocks forming the area of the pampean underground and the „Sierras pampeanas“ and lying under the cover of the conti- nental formations, are quite different in their lithology but as mentio- ned of a strong metamorphic habit. The age of the various members
‘is evidently also different. A recapitulation of earlier and later
_ views on this matter may be justified.
The first opinion, according to Stelzner (1876) was, that all the crystalline rocks outcropping in the „Sierras pampeanas“ belong
| to the precambrian. This author divides the whole complex into two principal lithological groups: 1) an inferior gneissic, intensely
‘granite intruded and of distinct abyssal character, and 2) a superior
schistose, of evidently sedimentary origin as a whole and only to
a limited extent intruded with granite. This Stelzners opinion
| must be considered quite a correct one, regarding the general litho-
logy and the relative position of the named complexes. But another question arises, when regarding the age of all these rocks.
According to Stelzners descriptions and sections from the Sierra de Famatina it is quite evident, that he considered all the quartzitic and argillitic schists of this range as of precambrian age, but that the relation of these rocks to the fossiliferous siluric strata of Potrero de los Angulos remained an undissolved problem. Concerning the discovery of silurian fossils Stelzner says, that he prefers not enter into discussion about the fact, that silurian rocks were found to the east of the Sierra de Famatina ridge, i.e. far away from the Silurian outcrops in the axis of the „Anticordillera“ (today named the „Precordillera“*). In order to dissolve this problem Stelzner had to few observations.
58 SIERRA DE UMANGO
Later on Bodenbender has more closely investigated the same — ridge and also many others of the „Sierras pampeanas“, and from ~ there concluded, that in the Sierra de Famatina and somewhere — else in the crystalline basement lower paleozoic sediments do occur. They are partly penetrated with granite, as in the case with the mentioned silurian strata in the Famatina mountain. On the eastern - side of the latter Bodenbender discovered cambrian slates, con- | taining Dictyonema and Staurograptus, and limestone banks with Obolus and Agnostus. In a report (1916) about the Nevado de Fama- tinaBodenbenders opinion is: The intrusive granitic body of the named mountain, now vastly exposed on the surface, contains great inclusions of various schists, probably all of cambrosilurian age (see the little sketch map, page 30). In some small parts of better preservation they contain fossils, but are usually metamor- phosed, to a quite high degree by the granitic body and its innu- merable apophyses. Also mingled rocks (migmatites) occur in lower horizons. The continental or „Paganzo“ strata (Gondwana and red beds) in the Famatina mountain are to be found only to a small extent, and Bodenbender says: „la propagaciön reducida del permiano-carboniano ........ a causa de su dislocacion y denuda- cion nos priva de poder observar eventuales intrusiones“. This statement shows, that Bodenbender had no firm opinion regar- ding the relation between the granitic intrusion and the covering »Paganzo“ strata. Further the author supposes, however, that the intrusion probably occurred in permian or already in carboniferous time, while the ,Paganzo“ seems to lie partly „conformably“ with the cambrosilurian schists. A such „conformity“ was observed by Stelzner too from Potrero de los Angulos. |
During a crossing of the Nevado de Famatina (Paso La Pampa— El Tocino) on the backward journey from the Sierra de Umango district the writer had some occasions to study the geological rela- tions. The many tectonically sunken remnants of red beds lie with basal detritic layers on the Famatina granitic rocks or on the por-
CRYSTALLINE BASEMENT 59
phyry complex, composing the higher parts of the ridges. The por- phyry series stands, as seems, in magmatic relation to the granite. At the western foot of the mountain lower „Paganzo“ (pale coloured “sandstones and conglomerates) lie on a granite ground, while on ‘the eastern side similar beds are downfaulted between the old por- | phyries and vertically tilted mica schists, the latter forming moun- ‘tains and ridges to the east from the Nevado. Nearer to Chilecito the granite again appears.
It seems quite probable, that in time between the consolidation of the Famatina batholith and accompanying porphyries, and the sedimentation of the continental strata there took place a great break in the succession of the geological records, i. e. a hiatus of erosion. The basal plane of the ,Paganzo“ series represents without doubt in general a deep geological horizon (cut) indicated by the predominance in the basement of coarse grained granites, molten and resorbed inclusions of rocks and also by the steep position of the mica schists. The observed ,concordance* mentioned by Stelz- ner and Bodenbender as existing between these schists and the „Paganzo“ strata may probably be explained as tectonical juxta- position, a common phenomenon also in the region of the Sierra de -Umango. I am consequently inclined to accept the old opinion of Stelzner, regarding the Famatina granite being of postsilurian age.
When regarding the other „Sierras pampeanas“ one may find, that there occur in many regions zones of sedimentary schists, limestones and conglomerates, apparently representing younger members, downfolded in the old precambrian gneisses and intruded schists. In the Sierra de Cordoba? for instance, we find amidst an old gneissic terrane a smaller remnant of coarse clastic layers, pres- sed conformably to the former and also many outcrops of folded limestone zones of evidently silurian habit. Also in the Sierra del Gigante, in the Sierra de los Llanos_and in the Sierra Pié de Palo
' 1G. Bodenbender: La Sierra de Cordoba. Anales del Ministerio de Agri- cultura de la Naciön. Secciön Geologia, Tomo I. N:o 2. Buenos Aires 1905.
60 SIERRA DE UMANGO | 4 there are many such younger zones of true sedimentary origin. + Keidel? and Rassmuss? have later on also expressed the opinion, that they may represent in some parts lower palaeozoic members, and that thence the hiatus under the covering permo-carboniferous strata may be less important than corresponding to the old opinion of Stelzner. | It is, of course, in many cases very difficult to determine, if the downfolded younger members really are of lower palaeozoic age. — From the province of Catamarca Rassmuss has reported a complex k of probable algonkian, separated from an older archaean body by a | marked unconformity (basal conglomerate). The younger complex i consists of quartzites and mica schists. In the sierras to the west 4 from Tucumän and further to the north there is a broad zone of chloritic and other schists, extending in the general N—S direction. The schists are of precambrian age, as shown by Keidel? from his investigations farther to the north, in the province of Salta, — where fossiliferous upper cambrian with a marked unconformity | lies over the strongly folded (and also mylonitized) complex of the : northern continuation of the same schists. ; Apart from these sedimentary zones the bulk of the complex of the „Sierras pampeanas“ seems to consist of gneissic and mingled © rocks together with extensive granitic bodies. The last named are ; partly of palaeozoic age, as shown by the fact, that they penetrate © the sedimentary rock members (limestones a. o.), whose lower pa- | laeozoic age is beyound doubt. Great gneissic complexes are met with in the Sierra de Cördoba, Sierra de San Luis and also in the ~ Sierra de Velazco. The coincidence of the orientation of all the schistose and gneissic | rocks of the „Sierras pampeanas“ (incl. the Famatina—Umango
1G. Bodenbender (1911). See the geological map. 2H. Keidel (1914). Page 674.
3 Juan Rassmuss (1916).
4H. Keidel (1914). Page 674.
CRYSTALLINE BASEMENT -
area) indicates, that the folding power at different times of pre- | cambrian and older palaeozoic periods was working almost in the same or in opposite directions. That fact increases the difficulty of ‘ recognizing structures of different ages in this region. | As a matter of fact, resulting from the foregoing treatment, the _ Famatina and the Umango schist complexes must be considered as older palaeozoic members, by downfolding incorporated in the old basement and strongly metamorphosed by subsequent granitic in- _trusion. The granitization belongs to a period, antedated to the time of the permian orogenetic phase. The lower Gondwana rests, as we have seen, with a great unconformity on the granite intruded schists. Thence, the Umango structure cannot, as Penck believes, be any continuation of the Precordilleran structure, but is an older one. The juxtaposition of these two structures is caused by later movements (the permian [Precordilleran] and also the tertiary [Cor- dilleran]). In the Umango area the deep geological level of the pampean underground and the ,Sierras pampeanas“ is thence brought side by side with the more supercrustal masses of the Precordilleran palaeozoic formations.
Because the position of the Umango area in the immediate front of the permian and tertiary andean folding zones, the area in question has played the part of a „foreland“ during at least two diastro- phic phases and has consequently also undergone tectonical influences, differing from these of the , Sierras pampeanas“ in general.
This circumstance makes the Umango area to an interesting field for studying mutual petrographical and tectonical relations, as the foregoing chapters may have illustrated.
he
me ay Wh aay
gt MN NAT fr Prins 2 Hi Sn Aral
N YA i ERRANG R x \
i
i
| | |
5
7
1
9 PART - ,POSTGRANITIC" CONTINENTAL DETRITIC FORMA-
TIONS. VOLGANIC ROCKS. TERTIARY TECTONICS. _ GEOMORPHOLOGY. MINERAL OCCURRENCES.
I A) THE BASIS OF SEDIMENTATION OF THE CONTINENTAL | STRATA.
The sub-Gondwana denudation.
The great variety of crystalline rocks composing the ,Umango structure" plays, as mentioned, the part of a basement supporting a mighty series of detritic, continental strata. Evidently a very great hiatus corresponds to the unconformity between these two lithological elements. It may therefore be assumed, that the denu-
dation which produced this unconformity, must have been conside- table as shown from the great areal exposition of the Famatina ba- _ tholith of today, with all its melted and mingled rocks originated _ by the abyssal injection phenomena.
_ The still visible remnants of the basis of sedimentation of the ~ continental strata in the Umango area and surroundings are pre- served only where the overlying strata have been recently re- moved by erosion, or where steep, crosscutting profiles are exposed. In all other parts the basal plane has been completely destroyed _ in consequence of later tectonical disturbances in the region, or it is concealed by still overlying strata. The following exposures may be mentioned: 1) Arroyo Blanquito, on the western side of the Cerro Villa Union to the south. Here the plane is steeply dipping _to the SE and overlaid with the lower Gondwana sediments. It
ee
2
appears also on the SE side of this mountain, in the Quebrada de los Lagares. Some leagues to the south in the high western profile _ of the Cerro Bola, described by Bodenbender (1911), the plane is not visible, because hidden beneath piedmont accumulations. 2) Cerro
~ ee
2 aoe
66 SIERRA DE UMANGO
Colorado and vicinity, on the eastern side of the Cerro Villa Union. — Here the plane is uplifted partly to a relatively high position and has a SE dip. In the through-going Quebrada Banda Florida far- ther to the north the plane forms the bottom of the quebrada and is overlaid with red sandstone. 3) Cerro Potrero Viejo, southern part. Here the plane is quite broken, but seems to dip in a sou- therly direction. Lower Gondwana lies over it with a thin basal conglomeratic layer. 4) Sierra de Safiogasta, southern part of the Nevado de Famatina. The plane rises to great heights from south- east in the direction of the summit of the Nevado, appearing as an even crest line. It is, hovewer, somewhat doubtful, if this plane can be regarded as an equivalent to the former, because the age of the overlying red sandstone still conserved in some parts of the crest is uncertain. |
B) THE CONTINENTAL SERIES.
Gondwana [Paganzo (La Rioja) stratal — upper meso- zoic and lower tertiary beds. Detritic sediments in perfect conformity.
According to geological maps! of the central western Argentina the areal distribution of the continental sediments, lying over the pampean — partly sunken and partly uplifted — crystalline basement, is varying in different parts of the region. While the outcrops of these clastic rocks in the more eastern parts as in the Sierra de Cordoba are insignificant, their extension increases gradually, when approaching the border of the Cordillera. The tectonical appea- trance of these series in the region of the „Sierras pampeanas“ is in general such: Along the border of the horst-blocks (the Sier- ras) they form an outward dipping fringe, an erosion remnant of
1 Ludw. Brackebusch: Mapa geolögico del interior de la Republica Argen- tina. Scale 1:1,000,000. Gotha 1891. G. Bodenbender (1911).
N
-
u: DETRITIC FORMATIONS 67
varying height of an once covering mantle, uplifted by the horst- movement. These fringes are only the visible edges of the vast sunken parts of the same formations, occupying the space between the Sierras and today completely covered with the more recent pampa accumulations. Nearer to the Andes the outcrops of the mantle are growing larger, because the tectonical throws are getting narrower and the pampean accumulations are decreasing in thick- ness in the same direction. Here the mantle-outcrops also climb to greater heights over the valley floors. Especially in the provinces of La Rioja and Catamarca these continental series are developed to great extent in the geological structure. Their most typical re- presentatives one will find in the former province incl. the Sierra
’ de Umango area, and therefore they have been named the La Rioja
strata. They correspond to the „Paganzo“ of Bodenbender.!
Farther to the south in the Precordilleran zone the extension of the continental strata, according to the geological map by Stap- penbeck (1910) comprising the area from Mendoza to Jachal, is less conspicuous, but still quite considerable.
The principal extension of the „Paganzo“ or La Rioja strata was studied and fixed on the map already 40 years ago by Ludw. Brackebusch.? The series in question was called by him „ferre- nos reticos", according to the earlier determinations by Geinitz (1876). They include also many volcanic sheets. The „rhaetic“ age, however, was in many cases considered by Brackebusch as doubtful, as may be seen from. the many notes of interrogation on the map. Later on Bodenbender (1896) divided the series more closely into „carboniferous“, „permian“, „triassic* and „rhae- tic" formations. This division was partly based on discoveries of plant remains in the strata. In a later paper by the same author (1911) one find the following scheme: the carboniferous, permian
1 (1904). See later on. 2 Ludw. Brackebusch, I. c. (1891).
68 SIERRA DE UMANGO
and triassic beds are united by him under the denomination „Pa- ganzo strata‘. In this group, lying generally in concordance, except the Precordilleran zone, and being wholly of continental character, 1 Bodenbender distinguished between three subdivisions of quite 4 different habit: „Piso“ I, II and III. The conformable hanging strata ° he considered as rhaetic, jurassic and cretaceous, the last named being in the subandean regions of coarse detritic, in extra-andean - (pampean) areas of calcareous habit. j
Regarding the Sierra de Umango area, a similar stratigraphical ° division will hardly be applicable, because of the many breaks in i the succession of the strata caused by tectonical tearings and also : because of the extreme scarceness of fossils. Stappenbeck al- I ready met with the same difficulties in the Precordillera (between Mendoza and Jachal). In similar circumstances worked also more recently W. Penck in the province of Catamarca, or nearer said in the region of the valley of Fiambala (1920). |
In the following pages there will be communicated a more detai- ° led description of all important occurrences of the continental strata, and then the writer makes an attempt to divide the whole series on the basis of lithological, stratigraphical and tectonical relations.
a) DESCRIPTION OF LOCALITIES. The Cerro Guandacol (The southern part of the Cerro Villa Union).
The most complete and most representative profile; through the lower part of the continental series is exposed here, especially on the abrupt western side of the mountain, some miles to the east of the village of Guandacol situated on the river La Troya.t The westfacing slope is due to a great fault, running at long distance ~ in a north-southerly direction along the eastern border of the valley of Rio La Troya (Rio Bermejo). The profile of the Cerro Guandacol was described at various times by Bodenben-
1 See the profile fig. 7.
ee NM
My Chis Mö
SS | S
FEY
UG
fa
DETRITIC FORMATIONS 69
der and also figured in a somewhat schematized manner (1897). Over the crystalline basement, which forms, as mentioned, a SE-dipping basal plane, there lies with the same dip (45°— 30°) a mighty series of continental Strata of different colours. This series continues along the southeastern side of the mountain (see later on), forming here a steep-dipping mantle. The whole series sinks rapidly beneath quaternary accumulations in the south, but far- ther in this direction it rises again in the great anticlinal upward bend of the Cerro Bola, as Bodenbender has already stated (1911).
The basement of the profile is com- posed of mica and amphibolite schists penetrated with red pegmatitic granite. The schist rocks have a slight north- ward dip contrasting with the steep inclination of the schists in all other parts of the Cerro Villa Union. This circumstance may be explained as a simple block tilting of the basement part lying under the sediment cover. The degree of the tilting is already evident from the dip of the strata.
Regarding the lithology of the strata,
UG: upper Gondwana red sandstones and coarse grained arcoses. V: volcanic agglomerates.
at the base there is a conglomeratic layer, composed of quite heavy boul- ders of angular forms from the same rocks, that occur in the basement. The
Fig. 7. A natural profile along the western side of the Cerro Guandacol (southern end of the Cerro Villa Union), seen from the Guandacol valley,
ome miles distant. To the extreme left (crosses) the crystalline basement. Ti: tillite. LG: lower Gondwana sandstones and shales. T: tuff layer.
70 ; SIERRA DE UMANGO
greenish matrix is here quite scarce. This conglomerate is well visible | in a locality called Arroyo Blanquito. The dip of the conglomeratic | layer is here very steep, nearly vertical, but this fact is due to a ~
local faulting of a later age than the general tilting. Upward the conglomerate changes its lithological habit thus that the size of the boulders decreases, while the relative quantity of the matrix is
growing mightier. The forms of the boulders change upwards to ~ more rounded ones. Finally the matrix is almost dominant, and —
a ,graywacke“ results without any conspicuous bedding. The rock contains small irregularly scattered boulders, and is quite well ce- mented: a polyhedrical jointing may be observed.
There is no doubt about the genesis of the ,graywacke“ rock, observed already by Bodenbender in another locality (see later on). The rock reminds one of the permian glacial boulder beds in the Precordilleran zone farther to the south, discovered by Keidel (1914). The lithological resemblance of the tillite described by Keidel in the classical profile at Jachal (I had also an opportunity
of seeing the rock) is very great. The coarse conglomerate at the —
base of the „graywacke“ in Arroyo Blanquito is clearly not water- worn. Although striae are not yet discovered, neither on the boul- ders nor on the basal plane, the conglomerate may be considered a moraine boulder bed.
Upward from the „graywacke“ in the profile (to the south) there fol- lows a sandstone of grayish colour, nearer the base alternating with a clayish slate. These rocks are overlaid with a similar slate as below and of considerable thickness. The slate is covered with a grayish sandstone.
I am inclined to regard the pale-coloured sandstones and slates —
near to the roof of the graywacke as a pelodite! or glacifluvial out- wash products from a time subsequent to the deposition of the
1 J.E. Woodworth: Geological Expedition to Brazil and Chile 1908—1909. Bull. of the Museum of Comp. Zool. Harvard Univ. Vol. LVI. N:o 1. Cambridge, Mass. 1912. Page 78.
+ SENSE en ge eee AA
DETRITIC FORMAT IONS 71
4 morainic matters. Somewhere in the sandstones and slates (pro- : bably more upward) Bodenbender has (1911) discovered badly preserved remains of plants, among which has been determined:
a
‘ Neuropteridium, sp.; Equisetites (?).
This series of pale-coloured sediments is capped by an almost
100 feet thick sheet of a peculiar chocolate-coloured, dense and unstratified rock, having the odour of humid clay. Microscopically the rock has been proved to be a pelitic tuff. Over this bed fol- lows in perfect conformity a mighty series of red and brown _ sandstones.
There is no doubt about the pale-coloured sediments below the tuff layer as corresponding to the lower Gondwana, i. e. to the Talchir-
Karharbari-horizons.
A series of red and reddish brown sandstones of the coarse arcose types succeeds over the brown tuff sheet with perfect conformity, partly almost psephitically developed and consisting mostly of an- gular grains of red feldspar and milkwhite quartz. The upper strata are all darker coloured and capped with volcanic agglomerate and ash layers.
To the south from the Cerro Villa Union rises, as has been men- tioned, the great upbend of the Cerro Bola, showing the same
| sequence of strata, but without the crystalline basement exposed. The profile here, forming an abrupt wall toward the Guandacol valley, was studied by Bodenbender (1897 and 1911). In these lower Gondwana layers he discovered in a carbonaceous layer the following plant remains:
Neuropteridium validum, Feistm.; Noeggerathiopsis, sp.
To the east the Gondwana layers are concordantly overlaid with thaetic and younger formations, dipping towards the Pagancillo low- land [Bodenbender (1911)].
Turning back to the Cerro Villa Union it may be observed, that the strata, described in the profile of the Cerro Guandacol, run with
72 SIERRA DE UMANGO
almost the same inclination, and direction along the whole southeas- 1
tern side of the Cerro Villa Union.
Looking from the Pagancillo lowland toward the mountain, the - steeply dipping cover appears as a gigantic „steel sheet“ capping —
the crystalline mass. A study of the succession of the strata in this cover appearing in profiles in the many crosscutting quebradas reveals, that several members from the Guandacol profile are lacking.
A good cut through the cover can be studied in the Quebrada de Los Lagares (fig. 8) situated to the northwest of the little ranch Aguada de los Burros (this locality lies on the path from Villa Union to Guandacol). Over the crystalline basement, here composed of a mica schist, lies with a dip of almost 45° a pale yellowish sand- stone, carrying a clayish slate in regular alternation. The basal conglomerate and the „graywacke“ lacks here. The yellowish sand- stone is succeeded by a red one of the same dip. The yellowish sandstone incl. the slaty interstratifications attains surely the thick- ness of two hundred meters. The red beds above contain some concordant sills of melaphyre, partly of spilitic development. A lon- gitudinal erosion valley separates the red layers from the immediate horst-flanc cover, the former standing out as long-crested foothills (visible on the map). The dark melaphyre sheets are discernible already at a long distance.
The thick, chocolate-brown tuff sheet from the Guandacol profile is not represented here. But the melaphyre corresponds surely to the same volcanic phase.
Further to the northeast one finds the same succession of strata as in the Quebrada de Los Lagares. In the vicinity of the mouth of the Quebrada Barreal (not named on the map), situated opposite the Cerro Punta Colorada, at the base there appears the same yel- lowish sandstone resting directly on the basement, here composed of granite and oi intruded and banded schists. The slaty inter- stratifications seem to be lacking. Over the yellowish lower Gondwana sandstone lies then a red sandstone of the same ap-
hrs = 2 ee = er TE Ära Den Eee eek SZ a ee we
DETRITIC FORMATIONS 13
pearance as that farther to the southwest. Regarding the tectonics
may be mentioned, that the yellowish sandstone upward changes
its dip forming a flat-lying cover over the crystalline bulk. It is
the only part of the area, where the Gondwana have been observed in an almost horisontal position resting zpon the basement’s upland surface.
_ Northward off this locality the Gondwana layers rapidly dissappear due to faults.
Fig. 8. A cut through the sediment mantle along the southeastern slope of the
Cerro Villa Union in the Quebrada de los Lagares. Lower Gondwana sandstone
and slate in regular alternation lying over the crystalline basement (mica schist). Looking south with the Cerro Bola in the distance.
. The region of the Cerro Punta Colorada — Las Ramaditas.
A continuation of the southeastern flanc-cover to the northeast forms in some degree the group of foothills, situated between the steep eastern side of the Cerro Villa Union and the valley of Vin- china. Here tlıe cover, however, has been broken into many small blocks and the outcrops of the strata show therefore a greater variety. Also the dips are more variable. This foothill-group consists of two essential parts, an exterior one, facing the eastern lowland, the Cerro Punta Colorada and its ramifications consisting of sandstones, and an interior one, Las Ramaditas, composed of
.74 SIERRA DE UMANGO
rocks of the crystalline ba- - sement. Among the crystal- line hills in the latter part there are many small rem- nants of sandstones too. The hills are separated from the Cerro Villa Union com- plex by a train of piedmont gravels, running north-south (fig. 9).
This broken ground may, as stated above, be regar- ded as a border tectonic to the great sunk area of the
The dark slopes
G | i fl
i
|
|
| |
|
|
|
the hills of Las Ramaditas.
occupying the whole underground under the gravel sheet. The unsigned _
wana, principally pale yeliowish sandstone.
3 Ne? Bu a Yang ta u ESR ee ea “ RR or past Br a Gen a a ARD 5
\
I
lj Il
i
ull
|
|
ee ee eee
pi Vinchina valley — Pagancillo
lowland. The strata in the Cerro Punta Colorada dip © toward the lowland, except with many small irregula- tities. The northern part forms a kind of mantle above the core of the crystalline ground of Las Ramaditas. The sediments of this area are well exposed in many profiles, especially to the south of Las Ramaditas. | They were examined repea- | tedly by Bodenbender ; (1896, 1897, 1911), princ- & pally at the foot of the Cerro Punta Colorada. The series here is more
The dotted parts (Gs) are red triassic beds, strata to the right (Gi) are of the lower Gond
Fig. 9. Looking south over the gravel plain between the eastern side of the Cerro Villa Union and
consist of old crystalline rocks,
DETRITIC FORMATIONS 15
complete than at the above described localities along the slope of the mountain. It shows, as shall be found, a remarkable similarity with the already described strata in the profile of the Cerro Guan- dacol. ot The sandstone ridge shows but few through-going quebradas, ; among which the most remarkable is the Quebrada Banda Florida. _ Another is the Quebrada Colorada, more to the south. In the for- mer gorge the sandstone ridge is cut across down to the crystal- line basement. In the interior the same immediate contact is visible in many localities, for instance at the northern foot of the Cerro Punta Colorada.
ib
| ; ;
Y
Fig. 10. The Cerro Punta Colorada from the north, showing the Gondwana strata dipping SE-ward from the crystalline basement (darker). In the fore- ground the subsequently dissected floodplain of Rio de los Noques. The stratigraphy here (fig. 10) is as follows:
. The lowermost members represent partly the „graywacke“-sand- stone, partly a conglomerate, described already by Bodenbender. Over the crystalline basement lies according to him a conglomerate with boulders of „schists“, ,graywacke“, ,carboniferous slate“ and of granitic rocks. The hanging strata belong (according to Bo- denbender) to „a gray psammite reminding of the graywacke“ (of the Cerro Guandacol). This conglomerate may thence not be a basal conglomerate like that of the Cerro Guandacol profile, but may represent an inter-Gondwana erosion-hiatus, while detritus pro-
76 SIERRA DE UMANGO
ducts of the lower Gondwana members are represented in the form - of boulders. The „grey psammite* of Bodenbender could not ; be the real „graywacke“ of glacial origion but must correspond to —
the upper, pale-coloured, psammitic layers of the lower Gondwana.
a
Then follows over the „gray psammite“ a red, apparently triassic j
sandstone. As to the writers studies of the strata in the profiles along the
Quebrada Banda Florida and the dry gorge of the Rio de los 4
Noques, running along the inside of the bordering sandstone mantle, the succession here is: over the crystalline rocks (quartzite schist)
Fig, 11. An erosion remnant of red Gondwana sandstone, lying on a" tilted crystalline basement block. Looking north from the Cerro Punta Colorada, eastern side of the Cerro Villa Union. {
lies a pale yellowish sandstone as the lowermost member. The li- thological character of this rock shows a great resemblance with the lower pale sandstones in the Cerro Guandacol profile. A distinct crosscurrent bedding reminds one much of the pelodite-sandstones in the classic profile at Jachal, containing small scattered boulders too. This sandstone is succeeded by a pale brown sandstone of psammitic character, overlaid with a mighty, very coarse conglome- rate of dark brown colours. The boulders of this conglomerate, well waterworn and lying stratified with sandy interstratifications, all consist of old crystalline rocks. Upward follows then a mighty
DETRITIC FORMATIONS | 17
series of red and brown sandstones of apparently triassic age. These last named strata form the very bulk of the Cerro Punta Colorada, as the name also shows. In the profile of the Quebrada Banda _ Florida only the red triassic strata are visible. They rest directly _ upon the crystalline basement, a circumstance met with in some ~ other parts of this area too (fig. 11). This lack of the lower Gondwana _ in some strongly limited parts may be explained as the result of an | incomplete inter-Gondwana erosion.
The „intermontane sandstone zone“.
A great variety of the continental strata is exposed in a long, strongly dislocated zone, running NNE—SSW through the whole country, and called here „the intermontane zone“, because it lies for a considerable part of its length between high mountain blocks and is intensely pressed together between them (see the geological map). The northern and southern extreme ends of the zone are somewhat difficult to fix. It may, however, be stated, that the zone to the southwest passes into the Precordilleran tertiary structure map- ped by Rassmuss, while to the north, according to the map of Brackebusch (1891) and W. Penck (1920), the same zone seems to continue in the mountains of the province of Catamarca, follo- wing the Valle Hermoso. This zone is a very remarkable tectonical
feature of the region. It represents, as shown later on, a subandean, tertiary, tectonical phenomenon, not found in the area of the true pampean sierras. In the present chapter, however, may be described only the lithological characters and the stratigraphical relations of the rocks.
The basal layers of this zone are visible only in two localities: 1) in Carrizal, situated between the Cerros de San Antonio and the Cerro Villa Union, and 2) at the Puesto Altillo NW from Maz. At Carrizal the relations are as follows: over an opturned mica schist complex with limestone layers lies a greenish conglomerate, com- posed of well rounded pebbles with soft green-coloured cement
78 SIERRA DE UMANGO
mass. A stratification may be observed in the conglomerate sho- wing, that the layers have turned vertically. To the west, that is upward in the profile, the conglomerate changes into a graywacke © like the rock from Agua Blanquita. All the pebbles in the conglo- | merate are composed of crystalline rocks. Probably the conglome- rate has a glacifluvial origin. Farther to the west no continuation : of the profile is seen, because this whole outcrop lies as an isola- ° ted sunk remnant in the crystalline basement. Some engl. miles - to the west there runs the eastern cuesta of the Cerros de San An- |
tonio, composed of strata of the higher continental series, but without connection with the graywacke at Carrizal.
The above-mentioned cuesta of the Cerros de San Antonio shows at the base, composed of the same mica schists as at Carrizal, a graywacke rock like the former (well visible in a locality to the NW of Carrizal). Here, however, the graywacke has been extre-
EI Färna Zn ben Lee ee NERE
mely crushed by the tertiary movements. This rock is succeeded \ upward in the bluff by reddish and brown coloured sandstones —
with gently westward dipping layers. These are not torn as the graywacke, a circumstance which may be explained only by way of assuming a special tectonical phenomenon (see later on).
In the other locality, where the bottom layers occur, at the Puesto Altillo, situated on the eastern border of the intermontane zone in the southern part of the Cerro Potrero Viejo, no graywacke nor conglomerate occurs, but a yellowish sandstone of lower Gondwana character. The sandstone lies immediately over the crystalline ba- sement, separated from it only by a thin sheet of a conglomeratic psephite. The sandstone contains regular interstratifications of a clayish slate. The thickness of this series is quite considerable, probably many hundred meters. The whole complex is strongly folded as consequence of the tertiary movements (in connection with the formation of the intermontane zone). The best exposures are visible not in the very zone, but aside from it, on the eastern slope of the southern part of the Cerro. On the very border, marked by a fault,
DETRITIC FORMATIONS 79
the sandstone and the slate of the zone have been intensely torn to a mylonitic mass of grayish colour. Following the fault farther northward in the direction of the Corral Martinez (lying on the eastern border of the zone) there always meet the same lower Gondwana rock series, downpressed and torn along the dislocation line. In some parts, however, the rocks have remained more un- broken and there the original lithology can be studied. In the vici-
aE:
Fig. 12. The „intermontane sandstone zone“ (dotted), strongly dislocated. In the background the Cerro Villa Union. Sketch from the eastern slope of the Sierra de Umango ridge, toward the SE. \
nity of the Corral Martinez the writer has observed a vertically in- clined part of a clayish slate, apparently a pelodite. The rock ex- poses (reminding of the quaternary banded clay in the Baltic basin of Fennoskandia) a conspicuous alternation of clayish and sandy layers in quick succession. In the sandy layers, (corresponding to the „summer“-deposits) I discovered carbonized remains of plants with an air of a Gondwana flora. Thinned and torn coal seams are outcropping some hundred meters to the south of this locality.
80 SIERRA DE UMANGO
>=
The coal is mingled with sandy matter. The yellowish sandstone © at Altillo contains on the bedding planes abundantly Cruziana- like figures. a
The pale yellowish sandstone and the clayish interstratifications — do occur only along the eastern border of the whole intermontane — zone. The principal mass of the filling sediments are younger sandstones of various colours (cinnabar-red, and also white, chocolate- brown, pale brown). The stratigraphical succession of all these — layers is quite difficult to state, because the tertiary movements have resulted in a number of strong folds and faults (see details under the tertiary tectonics). The elder members are, as seen, the pale yellowish sandstone and clayish slate with the coal seams. The crom-red sandstone of triassic(?) age, succeeding these beds in the slope of the Cerro Villa Union, seems to be lacking here, while some apparently younger brown beds develop a great thickness. Up- ward these gradually change into pale brown sandstone beds. These sandstones have a distinct continental littoral character with sun- cracks and ripplemarks on the bedding planes. The upper, paler variety contains some thinner conglomeratic layers. This sandstone occupies the western border of the zone and is overthrust from the west by the great Sierra de Umango fault block. On the very gli- ding plane there are, however, some small remnants of a crom-red sandstone, possibly the last remnants of the triassic (?) sandstone. The writer is inclined to consider the cinnabar-red sandstone to the east of the brown one as a younger one (cretaceous?).
An excellent cut through the whole breadth of the zone may be studied in a river gorge between the Jaguel bolson and the Vin- china valley. Here the ,intermontane zone“ lies outside the moun- tain blocks and plays the part of a ridge or a dividing crest-line between the two depressions. Its eastern flanc consists of a more than hundred meters high bluff of some sinuous trench and can. be followed from the northern end of the Cerro Potrero Viejo NNE- ward into the Valle Hermoso. The strata forming this bluff are
EIT IEW a a Got ES = er 5 i
i | DETRITIC FORMATIONS 81
almost exclusively brown sandstone with small outcrops of an un-
derlying red sandstone at the foot of the ridge. The brown layers lie in quite an undisturbed position in this profile, but soon they incline to the west, as shown in the transverse quebradas.
The river gorge between the Jaguel bolson and Vinchina (Rio Jaguel) reveals, going westward, the following profile: _ The brown sandstone layers, outcropping in the bluff at Vinchina in an almost horizontal position with red layers at the base, dip, as stated, westward with gradually increasing gradient. Simultane- - ously the chocolate-brown colour of the rock changes into a pale brown one, but of similar lithological character as the former. Further to the west, the consistence of the rock, however, becomes softer. The dip then attains its maximum (about 45° W) and re- mains so in the rest of the profile (see further on). _ This apparently quite uniform sedimentary series is overlaid with a mighty mass of surely younger strata of much softer consistence.
_ These lie in conformity over the older, but are separated from them
by a slight conglomeratic layer. The younger strata are of some different colours, pale salmon-red, or pale brownish, and often carry interstratifications of conglomeratic layers, attaining sometimes a considerable thickness. The consistence of the strata generally changes towards the west to greater softness. The dipping angle is always the same, 45° W. No doubt the soft series belong to young geological horizons (cretaceous-lower tertiary?), while the underlying sandstones are of the upper „Paganzo“ series. South off the described profile the soft strata are with strong unconformity overlapped by a young conglomerate (late tertiary?), described on pages 113—115. Also to the north the same overlap can be observed, as it seems, on a much greater scale.! |
The westward dipping strata of the zone disappear swiftly under
the modern accumulations of the Jaguel bolson.
1 This observation was made only by field-glass.
82 SIERRA DE UMANGO
At the southern end of the Sierra de Umango ridge the ,inter-
montane zone" shows a remarkable virgation. The western half of the zone bends into a sharp curve to the west along the sout- hern head of the ,crystalline* mountain block, while the eastern half continues in southerly direction to the Precordillera of Guanda-
a eS ae Zu
col, here passing over into the tertiary structure. Between these —
two branches lies the tectonical throw of Nacimientos -- Zapallar,
filled out by dislocated, young tertiary gravels. The eastern branch has entirely the same appearance as the just described ridge between the Vinchina valley and the Jaguel de-
Fig. 13. Tertiary arenaceons strata dipping W to the N from Sierra de Umango, at the E border of the Jaguel bolson. S. U. in the distance. Z = intermontane zone with folded sandstones. C= young conglomerate in horizontal position.
pression. The Cerros de San Antonio, as this southern ridge has been named, forms too a high bluff of several hundred meters to the
east running SSW and also almost S, at the same time bordering
the Zapallar depression from the Carrizal valley train to the east. The strata here show, like these in the northern ridge, a generally increasing dip towards the west, while the lithological character of the strata changes getting a softer consistence and a paler colour.
h å
The bluff on the eastern side shows the same brown, slightly dis-
turbed layers as in the ridge to the north of the mountains (the
stratigraphy of this bluff has been already more closely described ;
on page 81). A good cut through the whole ridge can be obtai-
i he ee ur Aue
ä
DETRITIC FORMAT IONS 83
fed by following the way from Zapallar to Guandacol along the crosscutting Quebrada de la Troya, reminding much of the above
_ named gorge of the Rio Jaguel in the northern ridge.
The strata in this profile dip under several angles. The river
_ upward, i. e. from the eastern mouth of the Quebrada, where the
great bluff runs, there are first the brown, slightly disturbed sand- stone layers. They dip gently westward, and at their base red layers are visible, exactly as at Vinchina. Along the river gorge upward one observes, that the dipping-angle of the brown layers gradually increases and that the position finally becomes vertical.
_ Further upward the strata are again less steeply inclined and simul-
_ taneously they change their lithological aspect and become softer
and more pale-coloured. The general strike is NE—NNE.
The westward bending branch of the zone is principally composed of brown sandstones with a substratum of crom-red sandstone. This last named is doubtless the lowermost horizon here, because it lies in a sedimentary contact directly upon the crystalline base- ment, as seen on the southern slope of the Cerro Cordobés. Upward in the profile the brown sandstone passes into a pale Drown one exactly as in the last described profile. These pale- coloured strata are outcropping on the inner border of the Naci- miento’s depression.
Summarizing the composition of this remarkable structural feature in the Umango area, called the ,intermontane zone“, may be stated, that this part of the continental sediment cover, strongly pressed during the tertiary diastrophism and downfolded between basement-
_ blocks, shows quite a uniform composition: chocolate-brown and
pale-brown sandstones dominate, the former occupying the eastern, the later the western part of the zone. Transitions exist between these two members. A crom-red sandstone occurs as small rem- nants in thinner zones or as basal layers of the series. Also a cinnabar-red and a white sandstone are seen in one locality. In
the northern part of the zone, where this has a considerable breadth,
84 SIERRA DE UMANGO
the upper, ‚pale sandstone is concordantly overlaid with soft youn- ger strata, not represented in the very “intermontane“ section of the zone. In this section a complicated tectonic is developed, with folds and overthrusts, but the northern and southern continuations show a simple dip toward the west, a monoclinal structure. The age of all these layers shall be discussed later on.
Small sunk sandstone masses in the basement.
Limited areas of sunk or down-thrusted sandstone remnants occur in many parts of the Umango area. Because these occurrences lack every relation to the other ones of the continental cover, it is a difficult task to fix their real stratigraphical position.
First is to be mentioned a somewhat narrow, but quite a long
zone of crom-red sandstone, running along the western foot of the main ridge of the Sierra de Umango and with a northeasterly trend -
crossing the same ridge in its lower, northern part. Finally the
zone flows together with the intermontane zone (Quebrada Salto). —
The red sandstone here is not only deeply downfaulted between steep walls of the crystalline basement, the strata stand also verti- cally in the greater part of the zone, or are laid into folds by a thrust movement from the west (see the tectonics). The lithological habit of the sandstone is very a similar one to that of the upper Gondwana red layers in the slopes of the Cerro Villa Union. Also
characteristic basic lava sheets are seen in these layers. The steeply inclined strata are discordantly overlapped with a coarse conglome- rate (description of the locality on pages 113—115) and this sediment conceals to a considerable extent the red layers. Because the most interesting tectonical phenomena are visible in the vicinity of the Puesto Pirca (at the western foot of the mountain ridge), the zone may be called in the following chapters „the Pirca zone“.
Other similar sandstone remnants in the form of longitudinal throws exist in the Quebrada Puesto Quemado, at the northern
end of the Cerro Cacho and also at the very western border of the
3 4 |
j
SVE NIER = OE Oe ee Te TOE
ERSTER EEE ARE FF
DETRITIC FORMATIONS | 85
Umango crystalline complex, on the watershed between the Que- brada Piedra Pintada and the Quebrada Cacho. In the former lo- cality an overthrust from the west is to be noted and also dynamic phenomena. Here the sandstone is of red colour and carries con-
_ cordant sheets of a basic tuff. Conglomeratic layers do also occur.
In the latter locality the red sandstone layers stand vertically bet- ween vertical crystalline schists on the eastern side and strata of
a graywacke and conglomerate series in similar position to the
west.
At the western entrance of the Paso del Cordobés (Quebrada del Cordobés) there are, going east, on the left hand a pale yellowish sandstone lying almost in a horizontal position against the moun- tain slope, consisting of old crystalline rocks. The sandstone has
purely white shades and contains conglomeratic interstratifications.
The whole character shows a similarity to the lower Gondwana. On the right hand there is a steep, high wall of old crystalline rocks, the sandstone remnant is thence evidently downsunk and preserved here in spite of the great absolute height of the locality.
At the eastern foot of, the Cerro Villa Union I observed on the trail from Villa Union to Maz a red sandstone immediately at the crystalline rock-wall, well exposed in the quebrada-profiles. This small outcrop is evidently the only visible remnant of a vast sunk sandstone area, occupying the floor between the named Cerro and the hills of Las Ramaditas and covered with the mountain gravel sheets. To which extension the lowland to the north from here may have a sandstone basement cannot be determined, because the alluvial fans from the mountain to the west cover all this ground,
The clastic complex west off the Umango area.
As has been already stated, the Umango-Cacho crystalline base- ment complex is bordered to the west by a great fault-line running north and separating it from an extensive series of steeply inclined
castic strata, composed, as observed, essentially of a graywacke-like
86 SIERRA DE UMANGO
sandstone in regular, alternation with quite a coarse conglomerate, containing well waterworn pebbles. At the fault-line itself these layers stand vertically, but to the south their position becomes more gentle-dipping with westerly inclination, while to the north, on ‘the western border of the Jaguel depression, the dip is toward the east. The longitudinal extension of these strata reaches pro- bably far into the Catamarca mountains, and southward at least to the northern head of the Precordillera of Guandacol.
There are following mountains composed of these rocks: from north to south the Cerro Punta Negra, the Cerros de Agua de la Piedra, the Cerro La Estaca and the eastern part of the Cerro Descubrimientos. The general strike of the strata is to the north, but a trend toward the northeast is observed in the northern part of the zone.
The zone marks the eastern border of the Cordillera in these
regions, except to the west of the Cerro Cacho, where the first
ridges built up by these series still belong to the topographical unity of the Cerro Cacho. Then follows to the west of these rid- ges the Pampa de Leoncito. Some small hills there rise over the gravel plains of this plateau and are composed of the clastic strata in question. The Cerro de Leoncito, bordering to the west the named pampa, is, however, differently composed, as is seen already from a greater distance. This region was studied by Rassmuss.
The sandstone-conglomerate complex lacks every lithological simi- larity with the continental series of the Umango area. It is consi- dered by Rassmuss of Gondwana age!. In any case the strata are of postgranitic age, because they dont show any influence of granitic intrusion, which outcrops in the immediate vicinity, amidst the Cerro Cacho crystalline schists. The strata have a purely clastic habit. The concordance betweem the schists and these strata is thence only a tectonical juxtaposition of tertiary age, pro- bably of a similar kind as the „conformities“ between schists and
1 Verbal communication.
\
a
DETRITIC FORMATIONS 87
the continental strata on) the Nevado de Famatina, mentioned by
_ Stelzner and Bodenbender (see the page 59).
The Pagancillo lowland and the Vinchina valley.
About the stratigraphy of the vast, sunk area of the Pagancillo lowland I have made but few personal observations. According to
Bodenbender (1911) the whole interior of this area is occupied _ by tertiary strata, lying over a basement of cretaceous sandstone,
outcropping along the borders of the depression, where upbending phenomena are seen. Also to the] north, in; the bottom of the
~ Vinchina valley, the cretaceous strata are exposed.
Good profiles can be studied along the course of the Rio Ber- mejo to the south from Maldonado and especially in the traversing
- quebrada of the Paso de Lamar. This water gap crosses the sand-
stone ridge, which forms the southern (southeastern) prolongation of the Cerro Bola and is known as the Lomas Coloradas with their highest point, the Cerro Colorado. To the south of the paso there rises another mountain, the Cerro Rajado, the northern head of a long sandstone cuesta, the Sierra del Valle Fertil. |
The profile in the Paso de Lamar begins in the west with pale greenish sandstone layers with numerous basic sills, according to Bodenbender (1911) of rhaetic age. The strata dip toward the east, but many irregularities may be observed, as flexures and a local döme-shaped uplift. Red sandstones of considerable thick- ness changing upward into brown layers, lie with a complete con- formity upon the rhaetic strata. These strata are considered by Bodenbender (1911) as of cretaceous (and jurassic?) age. The Rio Bermejo; upward, out of the paso, in the direction of Maldo- nado, the brown-red sandstone forms the whole underground, always dipping under the same angle to the east, soon covered with the old gravel sheets of the Pagancillo lowland. In the vicinity of the river bed the sandstone is well exposed, partly as isolated erosion remnants of peculiar shape. The slope of the Cerro Colorado to
88 SIERRA DE UMANGO
the west is barren and composed entirely of these sandstone 4 layers. | On his map from 1911 Bodenbender has indicated the exis- tence of tertiary strata along the course of the Rio Bermejo south from San Isidro (Paso de Isidro). I have, however, on this route seen only the above named red sandstone layers, exposed already in the described profile of the Paso de Lamar.
The writer has personelly traversed only the northern part of the Pagancillo lowland, between Villa Union and Puerto Alegre. ‘Here the road passes over a hilly landscape, composed entirely of soft younger strata. While the small meseta hills in the vicinity of the Villa Union consist of the red (cretaceous?) sandstone dipping gently eastward, the road leads further to the east into the above named soft-rock-terrane, known as the Lomas de Villa Union. The topography here has the character of „bad land“ and the strata are always pale yellowish or orange-coloured and of very soft consis- tence. They would probably have been levelled already by the erosion, if they did not carry a horizontal covering sheet of hardened diluvial gravels. The dip of the strata is constant, about 45° W. Conglomerate beds are frequent. The series corresponds to the Calchaqui formation. |
From the Puerto Alegre may be observed (with field glass), that the numerous terrace-remnants (high diluvial terraces) on the wes- tern slope of the Serrania de Famatina (the cuesta bordering the Pagancillo lowland to the east) are composed: of the same soft pale-coloured layers as in the Lomas. Here the dip is, according to the profiles of Bodenbender (1911), steep toward the west. The terrace planes therefore cut obliquely across the edges of the strata.
At Puerto Alegre in the erosion profiles of the dry current bottom is seen a series of vertically tilted layers of dark brown sandstone and conglomerate. Then follows to the east the red sandstones of the ,Paganzo“, disappearing toward the Famatina mountain abrubtly
DETRITIC FORMAT IONS 89
; on great fault walls. The strata are tilted here in many directions.
To the south of the Pagancillo lowland we meet with the great
sandstone terrane of the Campo de Talampaya, consisting, accor-
ding to Bodenbender’s map (1911), of cretaceous beds.
b) SUMMARY OF THE STRATIGRAPHY.
i Above is characterized the lithological habit and the mutual rela- tions of the continental strata, covering the old crystalline rocks.
| An attempt to arrange these strata into a chronological succession from the base to the top and a trial to separate some greater sedi- mentary divisions and to determine their age may be convenient. By, the almost complete absence of fossils this task will naturally obe only a very approximative one, based, as it is, on the lithologi- cal and tectonical peculiarities, and to some degree on the beha-
viour of the divisions to these in the surrounding regions.
The first question considering the age of the continental strata is that of the age of the basal plane. In other words: how old is the cutting plane in the Umango crystalline basement, supporting
all the clastic continental rocks and their effusives?
In the province of La Rioja aside from the Cordilleran chain
there is a considerable number of known occurrences of plant-
bearing lower Gondwana strata, forming the basal layers of the continental series and lying with a great unconformity over the steeply inclined and mostly granite intruded crystalline schists. The lower Gondwana occurrences, consisting of yellowish and gray
sandstones with coal seams and often’ with a basal conglomerate,
are: 1. Cerro Bola. 2. Cerro Villa Union (Cerro Guandacol, profile descr. page 68). ' 3. Nevado de Famatina: a) Quebrada de Tambillos. b) Carrizal. c) Potrero de los Angulos.
90 SIERRA DE UMANGO
4. Sierra de Safiogasta (Cuesta de Miranda).
5. Sierra de Velazco (Saladillo).
6. Sierra del Vilgo (Amanao).
7. Sierra de los Llanos. (Nacate, Malanzan, etc.).
According to the plant remains from several of these occurrences (Kurtz and Bodenbender) they belong to the Karharbari hori- zon of India or to the Ecca series of South Africa. The specimens belong to a typical Glossopteris flora (lower permian). Under the coal bearing sandstones lies, as said, in many parts a boulder bed of basal character, noted already by Bodenbender (1896) in the
Fig. 14. The tillite occurrence at the Agua Blanquita. Small circles signs: tillite. To the right in the background sandstones of the lower Gondwana. The white part to the left is the crystalline basement. Looking east.
Sierra de los Llanos, and since encountered in most of the above enumerated localities. The conglomerates are not yet more closely studied, therefore but little is known about their lithological charac- ter. Keidel is, however, of the opinion, that they are all of a glacial origin as the conglomerate of permian age more closely examined by him in the profile of Jachal in the Precordillera of San Juan (Keidel 1914). Probably they all correspond to the lower permian glaciation and are of the same age as the Talchir in India and the Dwyka conglomerate in South Africa. As found above the conglomerate and the graywacke outcropping in the pro- file of the Cerro Guandacol show a glacial (tillite) character too. Similar rocks also do occur in the Famatina mountain.
DETRITIC FORMATIONS 91
al
It may also be assumed, that the lowermost horizon of the con- ‘tinental series in this part of the Republic is represented by a lower “permian glacial formation. Probably this was deposited over a per- piety levelled surface, consisting of the crystalline schistose and q granitic rocks. Already above is stated, that the degradation process down to this surface apparently was going on, during a long lasting erosion period,! because abyssal rocks are exposed in the
sal
surface to a vast extent. This erosion work was naturally accompa- ‘nied by an uplifting movement of the ground. The plutonic meta- morphism, the results of which are now exposed in the surface, must thence have occurred a long time before the lower permian.
The basement cutting plane must thence at least be of pre-per- mian age.
In the Umango area there are remnants of the lower permian tillite, as said, only in the Cerro Guandacol profile and at Carrizal. ‘In all the other localities, where the continental strata are resting directly on the basement, either the yellowish sandstone lies lo- west, or the red triassic (or cretaceous?) one. The tillite beds (cong- lomerate and graywacke) have thence been extensively destroyed by erosion before the sedimentation of the red beds, or even be- fore the formation of the yellowish sandstone and shales of the lower Gondwana. These last named carrying in some parts re- mains of a Glossopteris-flora? (Cerro Bola, Cerro Guandacol, Corral Martinez) are in the Umango area generally of no great development, propably destroyed to a large extent by at least a pretriassic denu- dation. The mightiest layers do occur in the Cerro Guandacol profile, and a remarkable development is also visible on the southeastern slope of the Cerro Villa Union. No other occurrences of these
1 If this deeply reaching erosion proceeded in all parts of the pampean sierras as one phase or was interrupted in some regions by a period of diastrophism, is not closer investigated.
2 According to Kurtz the species Neuropteridium validum should be cha- racteristic of the lower Gondwana.
92 SIERRA DE UMANGO
layers are known except small remnants on the Cerro Punta Co- lorada and the Cerro Potrero Viejo. On the Famatina side there
are, however, several of this kind.
Over these strata follows without unconformity a series of crom- 8 red and upward redbrown sandstones with numerous conglomeratic 4
interstratifications. Already long ago this series has been regarded by
Bodenbender of triassic or Damuda age. The circumstance, that
these beds lie in some parts directly over the crystalline basement,
orogenetic movements!, as is evident from the conformity between these and the underlying pale-coloured sandstones and slates.
%
shows, that a pretriassic erosion took place, but without preceeding ©
The triassic red beds seems to have a considerable extension in — the Umango area, although it is difficult to state if the strata in -
some parts really belong to the triassic. As shall be seen, the upper layers in the continental series have often similar characte- ristics.
The development of the red beds on the Famatina side is a con- siderable one, appearing in most „grabens“ there and also on the crests up to great heights.
The fact that the conglomerate layers in i these triassic and younger strata carry boulders of the old crystalline rocks indicates, that judging from the thickness of the layers the pretriassic erosion reached the crystalline basement in some parts to a greater depth.
The red beds change upward into brown sandstones of similar consistence. They seem to be of a considerable thickness. It is interesting to find, that the brown strata occur only in the ,inter- montane zone“ to its whole length, but in no other part of the Umango area. For instance in the mighty succession of the Cerro Guandacol — Cerro Bola-profiles no brown beds are visible. But in the NNE continuation of the „intermontane zone“ in the mountains
1 It may be spoken later on about the influences of the permian andean move- ments here. They are in fact very insignificant.
DETRITIC FORMATIONS 93
Pen ck (1915) has really discovered similar beds (see later on).
4 The occurrence of the brown sandstone only in the intermontane | zone may be explained in the following way: At the time when the zone in question was downfolded or downpressed (in miocene time?), the continental series in the area was much more complete than now. The late tertiary erosion had not yet to any greater extent worn away the upper members of the series. The brown Strata extended still ower greater; areas. After these movements | the late tertiary erosion destroyed all the brown layers, except where they lie safe in the longitudinal throws.
It is impossible to express any opinion! about the age of the brown beds. The observation may be stated, that in the profile bet- ween Vinchina and Jaguel they change upward gradually into pale "brown sandstone like the rhaetic one in the Paso de Lamar, and that this later passes upward into softer, pale salmon and yel- lowish red sandy and conglomeratic beds of probably lower ter-
tiary age. The brown layers expose at their base a substratum of red sandstone. Also on the southern slope of the Cerro Cordobés the brown layers show downward red and upward pale brown sandstones in conformity. _
Surely rhaetic strata do occur in the Cerro Rajado at the western entrance to the Paso de Lamar, being here a northern continuation of the extensive occurrences in the Sierra del Valle Fertil, disco-
vered already by Stelzner-Geinitz [Bodenbender (1911)]. These strata contain remains of Thinnfeldia. The rhaetic forma- tion outcropping in the Paso de Lamar continues northward along the eastern side of the Cerro Bola. A farther extension of the rhae-
4 the west of the valley of Fiambala, province of Catamarca,
=
1 Penck (1915) mentions the occurrence of dark brown and redbrown sand- stone series in the western Catamarca mountains, resting with unconformity (quasi conformity) over the „Paganzo“ sandstones. Because they carry andesitic tuffs and conglomerates, Penck regards their age as tertiary — that of the lower horizons as probably cretaceous. Judging from the lithological habit of the brown beds in the Umango area, a tertiary age seems there to be quite low.
94 SIERRA DE UMANGO
tic strata in the Umango area seems to be doubtful; perhaps the pale-coloured members of the ,intermontane zone" are ofthe same age.
The many rhaetic occurrences indicated on the map by Bracke- busch (1891) from the Umango area are, as stated above, incor- rect.’ ö Over the rhaetic beds in the Paso de Lamar follows concordantly ; the upper red sandstone perhaps of jurassic — cretaceous age, very similar to the triassic one [Bodenb!ender (1911)]. The great litho- logical similarity with the triassic sandstones makes it very difficult | to judge about the age, especially regarding the small isolated ; remnants of such red sandstone, downfaulted into the crystalline ba- | sement. If the ,cretaceous* occurrences indicated by Bodenben- der (1911) from the eastern parts of the Cerro Villa Union are in reality all of this age, seems doubtful. For instance, the floor of the valley bottom in the vicinity of the Villa Union forms the direct continuation of the upbent strata at the eastern slope of the Cerro Punta Colorada and nevertheless the former are indica- ted as „cretaceous“, the latter as „Paganzo“. |
Still younger strata, belonging to the miocene, etc. Calchaqui-beds („estratos calchaquefios‘, named after great occurrences in the valley of Calchaqui in the province of Tucumän) have developed on quite a considerable scale in the Umango area, where they occupy pro- bably the greater part of the Pagancillo lowland. They also do occur in the tectonical depression of Zapallar to the west, so in the Ja- guel depression too, where they lie conformably over the older continental sandstones (profile Vinchina—Jaguel). The Calchaqui- strata consist of a series of soft, pale yellowish, pale salmon-red and white nuancering sandstone-marls and conglomerates.
A coarse gray conglomerate indicating a sudden change in the conditions of sedimentation does cover with a marked overlapping ~ unconformity these strata. This conglomerate is, however, of a small extension. All its pebbles consist of old crystalline rocks.
The probable stratigraphical subdivisions of the pre-
Sr En
Pliocene
DETRITIC FORMATIONS 95
Overlapping the whole series: Coarse torrential conglomerate (Puna gravels).
iUnconformity. Dislocations. Calchaqui-strata. Soft Sandstones, marls, conglomerates.
Dislocations.
Cretaceous Brown, red and redbrown sandstones. Crossing veins
Jurassic
Rhaetic
Triassic | (Damuda)
Permian
of basic volcanics.
Grayish and greenish sandstones and slates with coal seams and Thinnfeldia. Basic lava sills. {Pale brown sandstones (rhaetic too?). Red and upward brown sandstones (triassic too?) and conglomerate beds. Concordant sheets of amygdaloide and dense melaphyric rocks. Gypsiferous fissures in the sandstones. Suncracks and ripplemarks on the bedding planes.
Hiatus of erosion. Grayish sandstones. Clayish slates alternating with sand- stone beds and containing plant remains (Neuropteri- dium). Coal seams. |
(Karharbari) Grayish — yellowish sandstones with alternating slates.
(Talchir)
On the bedding planes of the sandstones Cruziana-like figures. Local hiatus of erosion. A graywacke-like rock with scattered small boulders. No Stratification (tillite). At the base a coarse conglome- rate bed. Great unconformity.
Old palaeo- Basement of metamorphic schistose and eruptive-injec-
ZOIC
tion rocks.
96 SIERRA DE UMANGO
C) THE PERMIAN OROGENETIC PHASE.
Dynamical influences in the Sierra de Umango schist area.
According to the investigations by Stappenbeck (1910) and the latter more detailed ones by Keidel (1914), the eastern bordering zone of the middle part of the Argentine Cordillera is tectonically composed of a pretriassic or permian folded structure, capped in a i great unconformity with the triassic red beds. These old folds are, — as Keidel has stated, of an „alpine“ character with ’nappes de i charriage’ etc., indicating the existence at that time of a tangential - orogenetic pressure from the west, possessing a much stronger cha- racter than later on during the tertiary diastrophism (see farther). This ancient folding zone, corresponding to the so called „Precor- | dillera“, extends from the vicinity of San Rafael in the south to the Sierra de Umango area in the north. The northern end of the fol- ding zone is formed by the „Precordillera of Guandacol“, recently ° investigated and mapped out by Rassmuss (1916). This part of i the chain rises to the west abruptly up from the Guandacol valley i floor. Northward it does not attain the Umango area, but is sepa- E rated from it by tertiary dislocations. Therefore the Umango crystal- — line rock complex does not come into contact with the Precordille- — ran structure. At the western border of the Umango area, where it is topographically connected with the Cordillera, no Precordilleran | structure appears, as seen above.
But it is evident, that the Umango area has at least in some degree played the part of a „Foreland“ to the Precordilleran chain. It re- mains to consider to what extent the tangential movements in the Guandacol Precordillera have influenced the Umango block.
In the Cerro Villa Union, situated opposite the Guandacol Precor- dillera, the Gondwana strata, as has already been described above, are disturbed only by the tertiary movements. Also in other parts of the area, where the lower Gondwana still remains in position preserved from the erosion, there are same circumstances.
Consequently the pressure phenomena characterized on the fore-
EPOSTA ~ = EN
VOLCANIC ROCKS 97
1 going pages, do not belong to this phase, but to earlier and later ones.
As has been stated, the lower Gondwana strata in the Umango area are frequently lacking where younger ones exist and where the red triassic or even the cretaceous (?) beds lie directly over the crystalline basement. This circumstance indicates an erosion hia- ‚tus, probably caused by the precordilleran (permian) tectonical influ- ences. During this time faulting movements seem to have occurred in the Umango area like those of the later tertiary time.
D) VOLCANIC ROCKS OF THE CONTINENTAL SERIES.
In the pampean sierras-area there are in the continental strata many occurrences of lava beds of different composition especially, as was stated already by Stelzner (1876—85), in the rhaetic beds. These volcanics are nearly all contemporaneous with the intervening strata, which is evident from the many amygdaloide rocks and the ash beds. — From the Umango area and the surrounding regions also many outcrops of similar volcanics are known, principally through Bodenbender’s investigations.
All the continental volcanics in the Umango area are of a very basic character. The most common types are melaphyre, porphyrite and spilite. Aphanitic varieties do also occur abundantly. The corres- ponding tuff layers are partly clayish and very decomposed, partly large-grained, or even agglomeratic.
But little can be stated about the occurrence of volcanic centres in the Umango area. The writer has discovered in some parts of the crystalline basement crosscutting veins of basic lava rocks, pro-
-bably corresponding to „furnish or feeding channels“ of the basic lava outpourings. At the eastern foot of the Cerro Villa Union there is seen a remnant of a volcanic neck, setting up through limestone. In the sandstone terranes there are sometimes accumulations of lava to a degree as to form doubtless volcanic centres. For instance the volcanic mass of the Cerro Rajado at the western entrance of
98 SIERRA DE UMANGO
the Paso de Lamar, described already by Bodenbender (1911), | has clearly furnished all the material of the numerous lava inter- {
already mentioned by Stelzner (l. c.).
The petrographical descriptions of the volcanics irom the pampean 4 sierras’ continental strata are not abundant. Stelzner (I.c.) says, that - these rocks are, when melaphyric, composed of a plagioclase and a purplish augite. According to this author ophitic structures are |
gancillo there is a vast accumulation of aphanitic and spilitic Ivan ¥
quite scarce. Olivine occurs sometimes, associated with picotite. — Magnetite is found frequently. The vesicles of the rocks, repre- — senting ancient superficial flows, are filled with calcite or viridite.
Sometimes they are also empty.
The petrographical data published by Siepert (1894) from the material of Brackebusch concern only to a small degree the Sierra de Umango area. Siepert divides the rock types into | „Diabases“ and „Melaphyres“. Regarding the geological age of © the rocks, the latter name is to be preferred. |
Specimens of the continental volcanic rocks, collected in all parts of the area investigated, may offer quite a complete picture of the whole effusive series, corresponding to the length of the continental — period in these regions. Besides a description of the concordant | beds of superficial lavas and tuff layers here follows a such of a | great number ofj effusive rocks, whose age in many cases cannot - be exactly fixed, i. e. quartz-porphyries, lamprophyric rocks, basalts and some leucocratic rocks of dike appearance. A description of -
at 3
the
the tertiary volcanics of some isolated occurrences will then follow.
a) CONTINENTAL" LAVA ROCKS.
The spilitic types are the most common among this class of rocks. - The dense mass is generally of a darkbrown colour, and calcite- or quartz-filled vesicles are dispersed therein as numberless white
“ we
Brg ae Sy
a
j
|
; | ;
VOLCANIC ROCKS 99
spots. There is also a porphyritic development containing con-
. spicuous plagioclase phenocrysts. The spilitic rocks occur princi- | pally in the triassic red beds, while the dense types are met with
in the rhaetic and younger strata.
The lava rocks have all been disturbed by the tertiary tectonical movements, but only exceptionally they are crushed.
The mineralogical composition of the different types of rocks do not reveal any special combination. The melaphyric facies is always bound to ophitic structure and richness in ferric oxyde. The ore
_ impregnation increases incidentally to such a degree as to obscure,
especially by pronounced weathering, the original mineralogical composition and structure. The plagioclase laths, however, are
often clear and lie thence in an almost opaque mesostasis of the ore pigment.
Description of types.
In the tilted red sandstone layers of upper Gondwana age along
the southeastern slope of the Cerro Villa Union already from a
greater distance there appear black, concordant lava sheets carved out by erosion. In handspecimen the rock is of a highly ferriferous, spilitic type. The abundance of weathered ferric oxyde conceals
_ the microscopical structure, except the plagioclase laths. These lie
isolated in the dark mass, producing a porphyritic structure. Albitic
twins are scarce, and the extinction is wandering. An incomplete
decomposition to sericite may be observed almost in all individuals. Some of the vesicles are filled with quartz or calcite.
Another sheet of the same sandstone series represents a dense, dark grayish melaphyre. The rock contains abundant plagioclase of imperfect lath-form in ophitic arrangement. The albitic twins are indistinct. Besides ferric oxyde, partly as secondary product,
there are no mafic minerals. Among weathering products calcite is
very abundant. Also small spots of chlorite are visible. The abun- dance of the feldspar produces the grayish colour of the rock.
100 SIERRA DE UMANGO
Many dark concordant lava sheets are met with in the Tambillos |
valley in the variegeted continental sediments in different grade
dislocated and occupying the whole valley floor. A melaphyre lava in —
the vicinity of the Puesto Tambillos shows microscopically a great abundance of ferric oxyde. This matter forms an opaque mesosta- sis between the feldspar laths. Cavities are scarce.
On both sides of the entrance to the Quebrada Arroyo La Totora southward off the Cerro Cordobés (opposite the Cerro Avalao) a series of red sandstone layers, dipping southward and reposing directly on the crystalline basement, contain in their lower part concordant sheets of an aphanitic melaphyre partly of amygdaloide development. The structure being a typical ophitic one shows a network of fine plagioclase laths, slightly albite twinned. The small-granulated augite is yellowish-grey; sometimes it is developed in larger, idiomorphic individuals. The interstices are filled with glass. Ore dust is quite abundant, partly associated with a gree- nish-yellowish serpentine and thence evidently an alteration product of olivine.
To the north of Nacimientos on the right hand of the way to Tam- billos rocks of red sandstone, dipping southward and forming a westward continuation of the mantle at the entrance to the Quebrada La Totora, contain a concordant basic sheet of a somewhat different aspect. The rock is dark brown and porphyritic with a dense me- sostasis. The distinct phenocrysts of plagioclase are pale greenish and isometric. The structure is closer andesitic than melaphyric. The feldspar phenocrysts show the zonar polygones well developed
with a maximum extinction angle (on cuts perpendicular to M and -
P) of + 25°, i. e. Ab,‚An,,. The groundmass consists of small feld- spar laths and minute ore grains, cemented by intensely pigmented
glass. The numerous cavities of the groundmass are filled with
calcite or viridite in radiating aggregates. Isolated grains (relics of augite) do occur too.
Near to the Pirca post in the above described small dislocation —
eM IEE CT Re ee TREE TREE
I en a RED ee Ne Pe Ta Se
hE Ee AE Te
‘ EN ” en RETTET
es
it ei
En VOLCANIC ROCKS 101
: zone the tectonically downfaulted red sandstone, running along the western foot of the Sierra de Umango ridge, contains a concordant | sheet of a highly amygdaloide, dark lava rock. All the well rounded cavities are filled with calcite. The amygdules, closely packed to- gether, lend to the rock in handspecimen an almost oolitic habit. _ The rock mass between the amygdules is a very fine grained, ophitic _melaphyre, very rich in ore minerals, which form an opaque mass between the minute feldspar laths. These are quite altered and pigmented by ore dust, and their optical characteristics could not be fixed. In the Paso de Lamar the series of rhaetic sandstones at the base of the profile contains a great number of basic lava intercalations. They are, according to Bodenbender (1911), of greenish melaphyres. There is also an abundance of the same rock in the rhaetic strata in the Cerro Mogote to the south. This lava rock was petrographi- cally described by Siepert (1894). The structure is a distinct por- phyritic one with isometric phenocrysts of plagioclase, showing partly a zonar structure. The groundmass contains small laths of plagio- clase, showing flow arrangement. Minute grains of iron oxide are scattered over this mass. A pseudomorph of calcite after augite is occasionally seen. Serpentine also appears, probably as alteration product of the original olivine. Regarding a more detailed mine- talogical description I refer to the quoted paper by Siepert.
In the cretaceous red layers, lying immediately over the rhaetic strata in the above named profile, some crosscutting dark eruptive veins are seen. In handspecimen this rock is dense but carries cavities. Microscopically it reveals a minute ophitic structure, composed of plagioclase laths, ore grains and calcite. Also opaque spots of ferric oxyde are. abundant, evidently of secondary origin as the cal- cite, and visible also in handspecimen. There are no mafic minerals
_ preserved. The cavities are filled with calcite. Seen between crossed nicols the rock is quite dark, probably depending on a not incon- siderable amount of glassy substance. The age of these cutting
102 SIERRA DE UMANGO
veins is perhaps cretaceous, judging from the development of amyg- 4
dules.
In the Quebrada Puesto Quemado, on the northern slope of ‘ the Cerro Cacho, there appears as intercalation in the strongly F
downpressed sandstone remnant a dark volcanic sheet, in hand- specimen difficult to determine. Microscopically it is of a typical mylonitic structure. The principal minerals are plagioclase, crushed
* into fragments, and ferric oxyde, filling the interstices between the —
fragments. The latter is probably always of secondary origin. Doubt- less, the original rock was a melaphyre, all mafic components of which having been transformed into mylonite and altered to a ferri-
pigmented mass. The habit of the sandstone containing the vol-
canic layer indicates also, that strong dynamical influences have overgone it.
To the north from Tambillos, the huge masses of volcanic rocks are generally of quite a different composition in comparison with the above described rock series. It is highly probable, that the former belong to a much later volcanic activity than the latter ones, and therefore their description may be postponed. However, in the vicinity of these volcanic masses there occur in the red sandstone beds belonging to the extensive series of the Tambillos valley floor some mighty lava beds of amygdaloide character, well visible on the path from the Puesto Tambillos up to the Puesto Umango.
b) „CONTINENTAL“ TUFF-LAYERS.
In the description of the profile at the western side of the Cerro Guandacol is made mention of the occurrence of a mighty tuff layer of dense habit and chocolate-brown colour, concordantly in- tercalated between the lower and the upper Gondwana series. The layer was already observed by Bodenbender (1911) and is indicated in the profile on page 69. In handspecimen the tuff rock is com- pletely dense and of no conspicuous stratification. The humid rock
smells clayish. Microscopically the rock seems to be very fine-
i a ne re nn a lee ET Zub ne Gm rar hen De es es a
mna wo
Ch se od is tee?
sv" = IM
ee ae
OSSE a ee ee ee
ee ae
es ee ae
Yo Ly ae” See ee
A N: =
2
R
;
y
N
4
VOLCANIC ROCKS 103
na although its intense pigmentation with an iron ore dust ~ conceals to some degree the structure of the rock. The greatest part of the grains seems to be quartz. A clayish product occurs probably in the pigmented mass. This product of chemical de- ' composition may be regarded as a result of a weathering of the rock under humid conditions, subsequent to the deposition of the : lower Gondwana layers. The tuff belongs to the first great volcanic _ phase during the continental period. Corresponding lava rocks are found farther to the northeast in a similar stratigraphical position - (see page 99). To the south of the southern end of the Cerro Villa Union there > follows the great upbend of the Cerro Bola with the strata ranged _ in the same order as in the Cerro Guandacol profile. According to Bodenbender it appear here many volcanic intercalations (spi- lites), but also tuff layers. All these dark intercalations are well visible on the mountain slopes to the northeast and north. In the syncline between the two mountains in the red sandstone _ beds there lie (near to Las Lajas) mighty volcanic agglomerates. _ These are mostly composed of angular fragments cemented by a dark mass of grayish colour (ash). The fragments consist of mela- phyre of porphyritic structure and also of quartz. Microscopically the melaphyre seems to be very ferriferous, having almost an opaque groundmass. From the last named locality some miles in the direction of Aguada de los Burros mighty series of red beds are outcropping | [according to Bodenbender (1911) of cretaceous (?) age] containing a concordant layer of a dark finegrained volcanic rock. In handspe- - cimen one detects small cavities. Studying the rock microscopically the principal part seems to be a glassy substance with numerous inclusions of angular quartz grains. The glass mass is very porous and rich in ferric oxyde, forming irregular spots. The quartz grains originate doubtless in the neighbouring sandstone, and the glassy tuff was thence evidently formed simultaneously with the sandstone.
104 SIERRA DE UMANGO
In the glassy mass there are also fragments of a dull feldspar
and of calcite. The calcite also fills the cavities in the rock.
E) CROSSCUTTING ERUPTIVE DIKES IN THE CRYSTALLINE —
BASEMENT.
In different localities of the Sierra de Umango crystalline area ;
there are many dikes mostly of porphyritic rocks and sometimes also greater volcanic masses cutting through the crystalline basement without visible relation to the covering sediments. Their age is therefore in many cases of a somewhat doubtful nature. In a few localities the dikes have been affected by the first tertiary mo-
vements. Generally these rocks are of a petrographical habit diffe- -
ring from the above described ,continental* lava rocks and show ;
a more pronounced hypabyssal behaviour.
Quartzporphyry dikes in Ihe Famatina assimilation zone.
On the western sides of the rugged crystalline foothills along -
the western slope of the Nevado de Famatina there appear a num- ber of straight running dikes of brown-red colour. The position
of them is remarkably flat. They show no relation to the old —
structure. In handspecimen the rock is dense, containing scattered -
minute phenocrysts of paler colour. Some quartz grains are also
megascopically visible. In some veins the dense groundmass has —
a flow structure. The rock is a typical quartzporphyry with a granophyric, sphaerolitic groundmass, composed of quartz and a
red-pigmented feldspar. The phenocrysts are to a greater extent —
of albitic character and thence give the rock a more keratophyric appearance. The quartz phenocrysts have nearly all strongly corro- ded contours. No signs of pressure are visible in the rock.
Lamprophyric dikes. In his paper on the geology of the Nevado de Famatina Boden- bender (1911) mentions, that ,lamprophyric dikes“ do occur in many
U
Er
VOLCANIC ROCKS 105
parts of this mountain. If this author may be correctly understood, rocks with name of „lamprophyres“ are partly younger crosscut- ting dikes, partly great inclusions in the batholith and thence elder than the granite. The latter types are probably identical with the above described amphibolite inclusions. There are seen but little of real crosscutting dikes of lamprophyres in this mountain, as far as I know.! The relations of these to the above described quartz porphyry (keratophyric) dikes have not been made clear by obser- vations. —
In the ,abyssal“ part of the Cerro Cordobés there has been disco-
vered a great number of lamprophyric dikes, independently cutting
across the old rock members. In these dikes the rock is always dense and of black greenish colour. There appears a jointing, causing a partition into angular blocks. Microscopically it reveals a porphy- ritic structure composed of small phenocrysts of plagioclase and a groundmass of granular amphibole of green colour, intermingled with small feldspar grains. The plagioclase phenocrysts are not more closely determinable, because they are intensely altered. The
-_ amphibole has the ordinary habit. A parallel orientation of minerals
in the rocks structure lacks completely, also every sign of pressure.
_ The groundmass is more of a hornfels nature. It seems quite pro-
bable, that the rock belongs to the malchitic type (Rosenbusch). It is remarkable, that the amphibole is more abundant than the plagioclase and thence also more abundant than in the ordinary amphibolite. |
There is in the limestone banks, occurring as concordant bands in the profiles of the Quebrada Cordobes (eastern slope of the Sierra de Umango ridge), a crosscutting lamprophyric dike, having
in handspecimen a dense habit. By microscope one can see, that
the structure is an ophitic one. The mafic mineral is an elongated augite, oriented in like divergent directions as the plagioclase laths.
1 From the collections of Bodenbender.
106 | SIERRA DE UMANGO
Also phenocrysts of this mineral are visible, with an idiomorphic
development. The augite laths are elongated in directions verti- cally to the b-axis, reminding of the limburgite form of pyroxene and also of the salite form. The maximal extinction is about 45? in sections normally to b. All these augite individuals carry a dar- ker border zone. The plagioclase laths are highly decomposed. In some parts there are pseudomorphs of fibrous serpentine, pro- pably after olivine. Iron ore occurs quite abundantly as spots or elongated grains, being partly of secondary origin too.
In the granite-free zone, consisting mostly of quartzitic schists exposed in the Quebrada Salto in the northern part of the eastern slope of the Sierra de Umango ridge, there is a concordant dark _ eruptive dike. The microscopical structure is of hypabyssal habit, but the rock is very decomposed by formation of secondary mine- tals (calcite, chlorite, etc.). The principal components are plagioclase and a mafic mineral, but the latter only preserved as pseudomorphs. The former shows allotriomorphic, the latter idiomorphic forms. The tracks of the cleavage in the latter lead to short prismatic amphibole.
In the higher part of the Cerro Cacho on its western slope a melanocratic dike cuts across the granite intruded schists.
The principal minerals of this rock are plagioclase, very altered, and a greenish amphibole with slight absorption (actinolite) and without pronounced orientation. Quartz is present to a small amount, representing the last product of consolidation. Secondary minerals are calcite and sericite. Apatite and magnetite are present as small individuals. The rock seems to belong to the vogesite — odinite series of the lamprophyric group. |
Basaltic dikes (melaphyres of the continental series?).
In different localities there are found basaltic dikes cutting across the crystalline schists. They are sometimes affected by the tertiary movements. Although the rocks lack every relation to the conti-
i
‘ig
-
VOLCAN IC ROCKS 107
3 nental sediments, they can probably be regarded as allied with the
volcanic series of the continental period, and may thence have
played the part of furnish-channels to the lava sheets of this time. The microstructure shows an ophitic development, although signs of
strong pressure often appear. The plagioclase has the labradorite characteristics. There is a pale purplish coloured augite which is
generally granulated. Secondary minerals are common. Quartz
- occurs in small quantities, likewise magnetite and pyrite.
F) TERTIARY (2?) VOLCANIC ROCKS.
From Tambillos to the north, up to the Umango highland ground, the passage goes through a hilly landscape of younger volcanics, situated on the southern border of the named upland plain (see the figure 15). These volcanics are not associated with the conti- nental beds as those in the cases above described (continental lava rocks), but form an isolated area, limited by faults from the surro- unding crystalline ground. The downwarping of this volcanic mass has also preserved it from destruction by the river erosion. As these rocks differ in geological and petrographical appearance from the continental volcanics, it seems probable that they belong to a much later volcanic activity and may be considered as contempora- neous with the tertiary andine outpourings of chiefly andesitic rocks.
The principal rock type of this volcanic series may be classified as a feldspar-porphyrite. In handspecimen it is a fine grained rock of a dark brown colour, carrying small feldspar phenocrysts with lustering cleavage surfaces. Microscopically there is an abundance of feldspar, not only as phenocrysts, but also as matrix component, where it is mingled with grains of iron ore. The feldspar pheno- crysts have idiomorphic contours and develop a core of plagio- clase and a shell of potash feldspar. The plagioclase is a sodic one. The groundmass feldspar develops minute tabular forms, di- vergently oriented. They do not show any albitic twins, but undula-
108 SIERRA DE UMANGO
ting extinction. Evidently it is a more acid (trachy-) andesitic |
rock. io : Another rock associated herewith, is a reddish brown porphyrite containing pale phenocrysts of feldspar. Microscopically it reveals - abundantly potash feldspar as phenocrysts and also as greater frag- | ments. The groundmass is composed of a glassy (2) substance 3 |
with inhomogeneous pigment of iron ore dust. The glassy residual
Fig 15. Lava and tuff accumulations (dark hills in the background) of probably tertiary age, seen from the eastern wall of the Tambillos valley. Looking north. The dömeshaped summit in the dis- tance is the Cerro Cacho. The promontory in the foreground is of the old crystalline basement.
The even slope is sheet flood gravels. mass contains occasionally cavities. There are also smaller frag- ments of quartz and the rock may be classified as a quartziferous syenitic vitrophyre.
Intercalated between these lava masses lie tuff layers, gene- rally of paler colour and well visible at a greater distance (cf. fig. 15). These layers dip all northward. Microscopically, the tuff contains abundantly grains of corroded quartz in an allotriomorphic mass of pigmented feldspar and small quartz grains, irregularly distributed. Another specimen of the tuffs shows a high degree
TERTIARY TECTON ICS 109
4 of pigmentation by ferric oxide dust, as may be confirmed micros- copically. Between crossed nicols the slide is only dull transpa-
rent, or nearly black. This mass seems to contain feldspar as principal component. Also larger feldspar individuals occur with a tabular development showing a perthitic structure. Regarding the „parallel extinction“ in cuts perpendicular to M the feldspar seems
| to be a potash one.
- 6) THE TERTIARY AND QUATERNARY GEOMORPHOLOGICAL
DEVELOPMENT.
a) A FIRST TERTIARY OROGENETIC PHASE. Folding tectonic in the Sierra de Umango area. As has been shown several times by Keidel (1910, 1914) a, o. the Argentine-Chilean Cordillera with its great) pile of mesozoic sedi- ments and volcanics was affected by a first folding process during
_ middle tertiary time, antecedent to a phase, when the present al- pine relief was formed. The two phases of diastrophism have been
distinguished one from another by way of morphological evidences.
While a strong tangential pressure from the west affected the andine geosyncline during the first phase, it may be assumed, that the ,Foreland“ to the east, that is, the western margin of the pam- pean sierras-area also came into the zone of the pressure influence. Studying the tertiary tectonics in the foreland, one really finds in
- many parts folding phenomena, caused by overthrusting from the
west. The first informations on these movements were brought by W. Penck (1915) from Catamarca, where the sandstones to the
west of the valley of Fiambalä are folded in part.! Later on
Bodenbender (1916) has mentioned similar phenomena also from the Nevado de Famatina. In the Sierra de Umango area there are
1 About the Famatina range (stretching northward to Catamarca) he says (page 649): „Junge, tertiäre Faltung charakterisiert die Famatinakette im Gegensatz zu den östlichen Gebirgszügen.“
110 SIERRA DE UMANGO
likewise folded sandstones, appearing here in the so called „inter- montane zone". This zone is, as said, limited to the west and to the east by gliding planes, the western one dipping 45° W and showing a fault-overthrust to the east. The zone border is here — overthrusted by the edge of the crystalline basement block of the | Sierra de Umango ridge.. All the sediments sau in this zone,
NINO MOM Y
nae
Pye aes 2 Ck NED aha DR
HEINE I: il IH: RYAN TIER BS en Ir
a a BEN ARSA EN ÅR
i
fat — Sa CAT aa N ie Fett: au
sai dit en
Ai
Fig. 16. Part of the ,intermontane zone“, to the NNE from the sector between the mountains. The
picture will show the various bendings of the continental strata. These are overlaid by the horizon-
tally lying young’ (tertiary) conglomerate (horisontally striated). In the foreground an edge of the moun- tain block of the Sierra de Umango. Looking east-northeast.
which runs between the mountain blocks of the Sierra de Umango ridge and the Cerro Potrero Viejo, are laid into strong folds, mostly upright ones. Also many faults do occur (see the figure 16). Out- ward the „intermontane“ section no folds are visible, but monoclinal dipping and bending of-the strata. Close to the eastern border of the ,intermontane zone“, on the southern end of the Cerro Potrero viejo, there is a strongly folded complex of alternating lower Gondwana sandstone and slate. Here
the folding may be regarded as an overthrust-phenomenon over
TERTIARY TECTONICS 111
the basement. The contact with the latter is partly visible and shows a mylonitic mass, a mixture of sandstone and slate fragments and also a schist-breccia. A similar thrusting dislocation may be observed on a tilted basement (the overthrust plane rises towards the east) at the northern end of the ,intermontane" section, at the northern head of the eastern mountain block, here called Cerro La Pampa (see the figure 19). The numerous fault lines, crossing the sand- stone zone in longitudinal directions, have caused many irregulari- ties in the succession and thickness of the strata.
un 100 m
Fig. 17. A West — East profile through the tertiary dislocation zone (the Pirca zone) along the western
foot of the Sierra de Umango ridge. Arrow I: di ection of the overthrust of old crystalline rocks over
red sandstones, folded by this movement. Arrow II: a later downfaulting of the late tertiary cong- lomerate. A=red sandstone, X = old crystalline rocks, C =late tertiary conglomerate.
Another, but smaller zone containing folded sandstones occurs, as have been mentioned, at the western foot of the Sierra de Umango ridge and has been called the ,Pirca zone". The red sandstone beds at Pirca are laid into relatively gentle folds, overthrust from the west by a faultblock of crystalline rocks, strongly crushed in the contact plane (see the profile 17). The amount of overthrust was, bowever, quite considerable, reminding of the ,lambeaux de charriage“ in the Alpes on a smaller scale. Blocks of old, crystal- line rocks lie on the red sandstone beds (see the profile fig. 18). Also in some other parts of the Umango area we find signs of overthrusting and folding from the west.
From the morphological circumstances which will be more closely exposed, may also be seen, that these movements really belong to
112 SIERRA DE UMANGO
an earlier phase than the below described vertical dislocations, by which the actual relief was formed. a
A ATNTIL ZOG CX Ö
N IT
Fig. 18. Profil through the western border of the Pirca zone; Over-thrust of a crystalline block (CX) over a red sandstone. Length of the block ab. 300 m. The arrow indicates the direction of the overthrust.
In the Precordillera of Guandacol these tertiary movements have caused many complicated tectonical features in the permian struc- ture, as is evident from the investigations by Rassmuss (1916).
The disturbances described above concern only the continental sediment-area series. But also the crystalline basement has been
Fig. 19. By -overthrust folded sandstone beds resting on the surface of the crystalline Basement (the thrust, plane). Northern end of the Cerro La Pampa and eastern border of the ,intermontane zone“. Looking south.
GEOMORPHOLOGY _ 113
Ree ver
affected by the tangential pressure, not only in the immediate vici- nity of the gliding planes, but also in the form of a regional ka-
taclasis more closely described in a foregoing chapter. These dy- namical deformations do not seem to extend eastward and outward the Umango area.
ob) A LEVELLING PHASE SUBSEQUENT TO THE TERTIARY FOLDING PROCESS. FORMATION OF A MOUNTAIN GRAVEL CONGLOMERATE.
The above treated tertiary tangential movements have intensely affected not only the crystalline schists and the granites, but also the sediment cover of the continental series. It has also been sta-
ted, that these movements belong to an older epoch than the almost
vertical dislocations now dissecting the Umango area in different
directions and causing the actual alpine relief. A closer proof of
this statement shall be made here and it will be clear, that an ero- sion phase lies between the two dislocation periods.
In the middle part of the Sierra de Umango ridge, on the very crest line, here at a somewhat lower level than farther southward there is a cap of a heavy conglomerate, extending downward on the slopes. The conglomerate rests partly on the crystalline base-
- ment, partly on the steeply inclined sandstone strata of the Pirca- »graben“ and of the ,intermontane“ sandstone zone.
The whole appearance of the conglomerate, the high position and the situation in a relatively sunken area of the actual ridge and also its great thickness indicates, that it must have formed in former times a widespread gravel cover, extending over a more continuous surface, but was brought to the actual position and extension by a later uplift and subsequent erosion. This erosion is still working with great power, and in a relatively short time all the conglomerate here will be worn away.
This conglomerate corresponds doubtless to an erosion phase following immediately after the folding-diastrophism. Before a clo- ser examination of this erosion and its results is made, it may be
114 SIERRA DE UMANGO
convenient to charaterize in short terms the lithological habit of the conglomerate.
The quantity of the boulders in relation to that of the cement is very considerable. The size of the boulders is always large, and badly rounded forms dominate. They consist entirely of old crys- talline rocks, gneisses, schists and granitic varieties of almost the
Fig. 20. Sunken erosion remnant of a young (tertiary) conglomerate, lying with great unconformity
over a zone of strongly dislocated, continental strata. The vertically striated part of the picture
signs conglomerate, the dark wills are basement crystalline rocks. The continental strata are buried.
— A graben in the high ridge of the Sierra de Umango. Looking south toward the Cumbre de Umango.
same kind as in many parts of the Umango crystalline area. The
conglomerate is clearly waterworn, and the cement seems to be a
calcareous sand (it is now considerably endurated). Stratification is not conspicuous in all parts, but somewhere a distinct one can be observed. The sediment is jointed to vertical cliffs of great height, a test of a considerable cementation. Interstratifications of sandy layers have not been observed. The whole formation is evidently an ancient mountain shingle deposit, formed during con- ditions of steeper gradients. The material does not seem to be of a distant origin.
This conglomerate lies but in slightly tilted position. It ends
ey wy ”
AR
ES Sei el nd a Zu > =
GEOMORPHOLOGY 115
toward the west and east by fault planes, but to the south, where the ground is gradually rising, it seems to thin out. The downwar- ping of the conglomerate here took place very probably simul-
| taneously with the other vertical dislocations of the area.
The conglomerate is worn away from every other part of the
region. But how is it with the basal plane of this accumulation?
This may still be seen in several parts, and especially, where the
levelling process has gone through the crystalline basement. Already in the introduction was mentioned, that all the high
mountains in this part of the province generally have a flat topped
Summit, almost corresponding to the heights of the upland surface
of the neighbouring Cordillera to the west. These summits might be the small remnants of a formerly levelled surface of wide exten- sion. It depends probably on the late age of the uplift of the pre- sent mountains and partly also on the greater resistence of the old crystalline rock-ground, that the remnants are still preserved. The former surface was surely worked out later than the folding of the continental sandstones, judging from the centre of the area, where
_ the mentioned conglomerate does occur. This may have been deposi-
ted on the same surface, whose remnants are preserved on the flat topped summits. It is evident from the character of the cong- lomerate, that this surface was not a „peneplain“ with a slow drainage. The general slope probably had quite a considerable gradient from a mountain region to the west, which was uplifted at same time as the the continental strata were folded. The great thickness of the conglomerate and the boulders always of old crystalline rocks show, that this erosion must have worked quite deeply into the basement ground. The detritic products of the continental sandstones, etc. have been carried further on and depo- sited on lower grounds. Some of the younger soft tertiary strata in the lowlands may be of that origin.
1 Probably this phase of erosion already destroyed older peneplains and erosion surfaces, such as the sub-Gondwana and the subcretaceous, except in localities, where the overlying strata have recently been removed by erosion.
116 SIERRA DE UMANGO
c) THE DEVELOPMENT OF THE ACTUAL RELIEF.
The network of the lines of vertical dislocations.
After the deposition of the above described coarse conglomerate, probably covering considerable areas, the ground of the area was affected by strong uplifting movements, causing, as known, the ac- tual alpine relief of the whole Cordilleran chain and of all the pampean sierras. These movements went in the Umango area as everywhere else along vertical or steeply dipping slide or fault pla- nes, running in different directions.
Along these lines of dislocation the cover of the continental sandstones sunk down with accompanying tilting of the strata and with formation of flexures or by a simple crosscutting of the layers,
so that their edge lies directly against the crystalline fault wall in
a more or less horisontal position. It is therefore quite simple to determine the position of the dislocation lines: these are represen- ted by the limit between crystalline floor ground and clastic rocks, in the topography marked by a more or less abrupt slope. On the accompanying map of the Umango area the principal lines have been fixed.
According to the statements above it is evident, that all the lo- wer areas really must be occupied by sunk parts of the continental sediments, while the greater elevations are composed of the crys- talline ground. ‚These elevations are but very exceptionally capped by the uplifted parts of the sediments, since almost all of them have been destroyed by erosion.
A short description of sunk parts of the sediment cover, that is, the tectonical valleys and lowlands, will follow in the next pages.
Description of the sunk (down faulted)} areas. The Vinchina valley.
As mentioned, the Sierra de Umango topographical complex is very markedly separated from the Famatina ridge to the east by the deep broad and long Vinchina valley running from the moun-
UR nun Zum m ee na
GEOMORPHOLOGY 117
tains of the Catamarca province (in the north) southward with an-
increasing breadth and ending in the Pagancillo lowland. The bot- tom of this great valley consists of sandstones of the continental series, mostly of red colours, to the south, however, overlaid with younger, tertiary strata, attaining a great extension in the Pagan-
- cillo lowland. It is evident that these substrata of sandstones ge- _nerally must be of no great thickness, because in the middle part of
the valley, in the vicinity of Villa Castelli, a great number of smaller
crystalline hills occurs, covering quite a considerable area (Cerro
_ Asperecito, etc.) The walls of the valley almost expose the crystal-
line ground. North from the Sierra de Umango ridge the western wall is, however, a sandstone ridge, separating the valley from the
Jaguel bolson. More to the south the same side is formed by the
sandstone hills of the Cerro Punta Colorada. The eastern wall, the high slope of the Famatina mountain, shows in the foreground a series of smaller crystalline hills extending from Villa Castelli some lea- gues to the south.
The tectonical type of the valley is a longitudinal throw or „graben“.
- There; have been minor faults along the chief lines of dislocation,
as is seen from the occurrence of several smaller faultblocks of longitudinal extension, now appearing as foothills. The strata of the Cerro Punta Colorada are tilted in direction of the valley and show a continuation into the bottom layers. The Cerro Punta Colorada is thence formed by a bending up of the substrata of the valley by the rising of the basement here in the form of a separated fault block. Or
more exactly, here is a whole system of small blocks, as stated
before.
The Pagancillo lowland.
The lowland forms an immediate continuation of the Vinchina valley. It is limited to the east by a southern prolongation of the great western flanc-fault of the Famatina mts, here cutting across an uplifted table of the continental sandstones and forming a bluff or
118 SIERRA DE UMANGO
„euesta“ facing west. This dislocation was mentioned by Boden bender (1911). To the west the lowland is bordered by the up- —
bent edges of the same strata, here forming the ridges of Lomas Coloradas (Cerro Colorado) and Cerro Bola. In the bottom of the lowland the strata seem to lie almost horisontally.
Simultaneously with the sinking down of the Pagancillo lowland
TLS ee
al oe
many disturbances occurred: between the Cerro Villa Union and the j Cerro Bola, causing the marked downfold of the strata now sepa- -
rating the two mountains. The axis of this synclinale runs NE. —
But while the strata on the eastern side of the Cerro Bola dip to the east, a great bend of the strata may be observed when regar- ding the mountain from the northern side.
The Guandacol valley.
The most extensive lowland in these regions, the valley of Guan- dacol or of the Rio Bermejo with its continuation far to the south, is the dividing line between the Precordilleran chain and the Cerro Villa Union — Cerro Bola. I have studied only the eastern wall of the valley, as the western wall was mapped by Rassmuss. It is interesting to see, that the great fault line, which forms the eastern wall, is younger than the dislocations of the Cerro Villa Union on its southeastern and eastern sides. In the southern edge of the last named mountain the SE-ward tilted strata of the Gondwana series are vertically cut across to the west by the above named great fault line. The downfold between the Cerro Villa Union and the Cerro Bola is consequently older than the Guandacol valley. |
The young fault line along the eastern wall of the Guandacol valley
was already described by Bodenbender (1894). It is a very |
long one forming the western border of the Cerro Villa Union, the Cerro Bola, the Cerro Rajado and the Sierra del Valle Fertil more to the south [see the map by Bodenbender (1911)]. Toward the north the valley is rapidly getting narrower and the bottom rises.
a GEOMORPHOLOGY 119
4 The narrowing is due to an intervening ridge of sandstones, the Cerro San Antonio, running north (see the map!). £ 4 : | i
The tectonical valleys of Tambillos-Guandacolinos.
These valleys may be regarded as a tectonical continuation of i the Guandacol valley to the NW, separated from it by the named | * sandstone tidge of Cerro San Antonio. The Precordilleran border still - forms the western wall, here a more abrupt one. To the east rise the high slopes of the Umango mountains. On this side there are _ observed vertical gliding planes with tectonical striae. The valley of Tambillos is filled with sandstones of the continental series well exposed by the erosion. They are mostly dipping westward. The Guandacolinos valley more to the northwest lies higher than the Tambillos valley and thence the bottom sandstones here seem to have been worn away by erosion. Crystalline hills are therefore not unusual here. This last named valley cuts obliquely across the Umango-schist-structure. It ends where the Umango complex gains the Cordilleran border.
The Jaguel bolson.
It has already been stated, that the Umango structure is separa- ted to the north from the similar structure in the Catamarca moun- tains by the depression of the Jaguel bolson, where the old rocks are covered with a mighty formation of sandstones and recent allu- vions. The Jaguel bolson is a typical kettle-formed „graben“, sur- rounded on all sides by mountain walls, except to the east, where a low sandstone ridge rises, which forms the northern continuation of „the intermontane sandstone zone“. To the west and east the slopes of the walls consist of the continental sandstones, dipping inward. To the south there are no sandstones visible as flanc- cover, but the old Umango structure is here cut across, dissappea- ring rapidly beneath the young accumulations. The same peculiari- ties seem to be dominant also on the northern border of the bolson.
120 SIERRA DE UMANGO
The sinking down of this bolson evolved before the high seated —
conglomerate in the Sierra de Umango ridge was deposited, because
this conglomerate lies unconformably over the tilted strata of the ©
eastern border of the bolson (or exactly, over their southern conti- nuation). The strata were naturally tilted up together with the for- mation of the bolson. |
The above named conglomerate, situated on the ridge of the Sierra de Umango, is, as has been stated, preserved in a smaller downfaulted area, which seems to be of a later origin than the valley-iorming dislocations. |
The nearly smooth ground, which forms a kind of highland plain
between the Cerro Cacho and the Sierra de Umango ridge, may be considered as a relatively sunken part of ihe older tertiary le- velled surface. No remnants of the sediment cover are preserved in this area, and that depends naturally on the relatively high po- sition of the plain, which slopes rapidly down to the north and south.‘
The crystalline horst-blocks of the Umango area have in many respects the same character as the pampean sierras, formed as they are by similar dislocations. But the conspicuous phenomenon of tilted blocks, as generally developed in ihe last named sierras,! is hardly seen in the Umango area. The Famatina mountain blocks seem, however, to follow this rule, according to which the western borders of the mountains have been more uplifted. Also the Cerro Villa Union has a tendency to a similar behaviour, although worse developed, while an upland plain is but little preserved. The western slope is, however, higher and steeper than the eastern. Tilting does not occur in the principal part of the Umango mountains. They form a part of the Cordilleran late tertiary uplift zone, and in this respect the area lies aside from the pampean sierras-region, while the Nevado de Famatina may already belong to the last named.
1 Juan Rassmuss: Rasgos geolögicos generales de las sierras pampeanas. —
(1916).
| GEOMORPHOLOGY FE KN DA!
Id) THE POSTPLIOCENE EROSION AND ACCUMULATION.
General conditions of erosion.
With the great late tertiary uplift-movements a new geogıaphical
cycle began in the andean regions. From the present morphologi-
cal forms in the Umango area may be concluded, that this cycle _ must still be in a little advanced stage. The elevations are to their extension still determined by the fault lines. This statement is app- licable even to the most intensely faulted grounds, where every hill is a little horst. The youthful stage of the erosion is also manifes- ted by the deep and narrow profiles of the valleys, the rapidly de- scending erosion-curves on the slopes and also by the still pre- served relics of upland plains.
The general topographical situation of the Sierra de Umango area in the interior part of the Republic may be characterized as follows: The area lies aside from the pampean plain and in a high position with respect to the local baselevel. The mountairs form, as said, really only a part of the Cordilleran bulk, except some more isolated parts (Cerro Villa Union, a. 0.).
Examining indeed the erosion-forms of the Umango area, be- longing no doubt to the physiographical province of the pam- pean sierras with their very rough and dissected surface, one reveals the dependency of the erosion in the last named sierras as well as in the Umango mountains upon the possibility to attack ihe more or less isolated blocks from all sides, and upon the existing great level differences. Westward again in the Cordillera the vast high- land-area misses deep dissected valleys, and thence the linear ero- sion has but slightly progressed. The Cordilleran upland surface is still preserved over wide areas. To these belongs the part lying immediately to the west off the Umango area and contrasting with the later in a very conspicuous manner.
The principal outlet-course for the whole drainage in the Umango area, leading down to the pampean plain, is represented by the
122 SIERRA DE UMANGO
Rio Bermejo, which follows in its upper part the Guandacol valley and runs then southward along the western border of the Sierra de la Huerta between this mountain and the Sierra Pié de Palo. Fur- ther to the south the river joins the Rio Desagiiadero. Here the plain lies 457 m over the sea. The last named river flows then further southward, ending in a wide depression without outflow, in the Gobernaciön de La Pampa.
At present time the Sierra de Umango has consequently no drai- nage to the sea, in this respect being of the same nature as the whole interior part of the Republic between Tucumän in the north and the headwaters of the Rio Colorado in the south.
The more distant bases of erosion are, however, of no importance for the development of the land-sculpture in the Umango area. Here may only be considered the local bases, each of these being in arid conditions and of great influence upon the forming of the sculpture.
The lowermost local baselevel of the Umango area is represen- ted by the bottom of the Guandacol valley. Other somewhat hig- her levels are the Pagancillo and the Jaguel bolson floors. The absolute heights of the different plains are as follows:
Valley of Girandacols:. Yes, 910—950 m. Pasancillo lowland N... ee 1190 m. Valleyeok Vinchina oi. mer 1260 —1400 m. JA SUSAN OLSON So sen Nee 1844 m. VaileyeotePambillos so nase. 2225 m.
At present the erosion is controlled by an arid climate,! which seems to have prevailed during the whole quaternary and recent periods. The development of the landsculpture in desert regions has been explained by W. M. Davis.” According to W. M. Da- vis, the enumerated local erosion bases cause a corresponding
1G. Davis; El clima de la Repüblica Argentina. Buenos Aires 1909. Page 73. 2 W. M. Davis: The geographical cycle in an arid climate. The Journal of Geology. July-Aug. 1915.
PES USB BR.
GEOMORPHOLOGY 123
å number of independent centripetal drainage systems with heavy accumulations at the base. In the Umango area the different base- ‚levels communicate by small streams cutting across the dividing ridges, but their scarcity of water prohibites any sudden levelling between the different bases.
According to a paper recently issued by Keidel! the mountains in the northwestern parts of the Argentine Republic got their prin-
cipal relief already in preglacial times. In the formerly glaciated areas, it may be observed fhat the mantle of glacial deposits covers all the main forms. The piedmont-stream-deposits are consequently chiefly of preglacial age. The preglacial climate was dry as the present.
In the Umango area glacial deposits are completely lacking and ‚cannot furnish any information in this respect. It is, however, quite certain, that the principal drainage lines here, viz the main erosion lines are of old age. This is evident on regarding the giant gravel cones radiating from the mouth of every such valley. The cones are generally uplifted and dissected by younger erosion (cp. fig. 23).
1) A first erosion phase.
The late tertiary tectonical relief was in its initial stage probably built up as follows: all the great blocks were composed of the basement rocks and flanked by the continental mantle of sediments, which generally dips outward from the block cores. At an earlier time this mantle probably covered also the very summit of the blocks. The young tertiary conglomerate capped probably all the summits.
The present strong relief was brought out chiefly by the carving out of the hard crystalline blocks as result of the destruction of the continental strata in the depressions.
1 Hans Keidel: Ueber den Anteil der quartären Klimaschwankungen an der Gestaltung der Gebirgsoberfläche in dem Trockengebiete der mittleren und nörd- lichen argentinischen Anden. Compte Rendu. XII. Session. Congr. Géol. Intern. Canada 1913. Ottawa 1914. Pages 757—769. ‘
124 SIERRA DE UMANGO
During the first phase of erosion, a mighty cover of coarse cong- lomerate (the late tertiary) was really destroyed to an extensive degree which is indicated here and there by visible remnants of quaternary shingle, composed of the same boulders as the young
tertiary conglomerate, thus representing a re-deposition of tertiary- —
material. These quaternary shingle deposits were afterward destroyed by subrecent erosion to a considerable degree. They occur at loca- lities, where there are signs of dislocation of later date.
Such a remnant of quaternary shingle, afterward dislocated, is visible at the southern end of the Cerro Potrero Viejo, at the eastern border of the ,intermontane sandstone zone“. A recent river gorge
2) >
x G A
Fig. 21. An E—W profile from the vicinity of Altillo, E border of the „intermontane zone", showing tertiary and diluvial dislocations. CD=diluvial conglomerate. X = micaschist. G= Gondwana sandstone. A=red sandstone. C=tertiary conglomerate. A= red sandstone (cretaceous ?) is carved out through the gravel and also through the older strata to the west, against which the gravel has been dislocated (see the profile, fig. 21). The gravel being distinctly bedded with much sandy material has been laid into a flexure. Upward the profile is limited by a gentle eastward dipping surface of a young sheetflood. The quaternary dislocation had here a direction parallel to the tertiary of the intermontane zone, as the profile figure shows. Another remnant of diluvial gravels, to some higher degree har- dened to conglomerate, was found in the valley bottom of the Rio La Troja, at the western foot of the.Cerro Villa Union. The gravels form here a mighty vertical wall facing the eastern margin of the river, but slope gently to the east, toward the recent mountain fans. In the river cuts the gravel shows almost horizontal layers.
Ba ee TS ee ee
GEOMORPHOLOGY 125
The lateral erosion of the Rio La Troya has probably extensively destroyed this gravel formation and the levelled part has been completely covered with modern alluvions. At Carrizal the gravel is limited by a fault.
It is very probable, that such a gravel also occurs in the Jaguel bolson (in the north), but the recent alluvial cover does not permit any observations, except in one place, in the immediate vicinity of the settlement of Jaguel, where there rise some low hills over the plain, which are composed of similar gravels, but slightly hardened and not much dislocated. One may assume, that a great part of the filling masses of the bolson consists of the same gravels, as the bolson was re-deepened also in posttertiary time.
After the destruction of the young tertiary conglomerate cover the erosion pursued in the continental sandstones lying on the ba- sement block sides. This denudation process appeared in a suc- cessive formation of sheet flood terraces, on lower and lower levels.
2) Sheet Flood erosion.
This characteristic kind of mountain degradation was more closely explained by Mc Gee.! The author states with regard to the ge- neral conditions of sheet flooding as follows: „The first requisite for typical sheet flooding is a precipitation so rapid, as to exceed immediate absorption of the dry earth and immediate evaporation in the dry air. An attendant condition is, that the precipitation shall be rapid. A third condition is, that the soil shall be readily pervious only in limited degree. or to limited depths. And this condition is met with on the lightly veneered baselevels adja- cent to the mountains, where the mantle only is porous and the underrocks sound and hard; it is not met with in the deeper central portions of the valleys, where the permeable sands are of conside- rable depth.“ The gradient of the slope must be about 300 feet
1 Mc Gee: Sheet flood erosion. Bulletin of the Geological Society of America Vol. VIII. 1897. Pages 87—112.
126 ; SIERRA DE UMANGO
per mile. Further the quoted author states: , The sheetflood is characterized by a tendency to spread widely in relatively narrow sheets (of gravel loaded water). In a general way, streams prevail in humid regions, sheet floods in arid regions, though streams occur locally also in arid lands, while it seems probable, that sheet floods | occur under certain conditions in nearly all lands, howsoever humid.“ —
Almost all the above enumerated conditions prevail in the Sierra de Umango area, and the sheet flood phenomenon has therefore
Fig. 22. High destructive, diluvial (?) terraces, carved out in red sandstones. Southeastern side of the Nevado de Famatina at Puerto Alegre. Looking north. The high ridges in the background are granite. The ranch lies on a subrecent sheetflood surface. here a very typical development. Rapid precipitations occur in summertime in the form of the so called ,crecientes«. The soil is in the shallowest underground already quite hard, consisting of the
endurated continental sandstones.
The same conditions for sheet flooding have prevailed down to quite recent times during the successive degradation of the sand- stone mantle oi the crystalline block sides. The whole process was merely a formation of lower and lower, gentle sloping terrace pla- nes, a kind of destructive terraces, carved out in the sandstone mass.
Very few older sheet flood terraces (situated on higher levels) are yet visible. In the ,intermontane zone“ there are in some parts
GEOMORPHOLOGY 127
along the eastern slope of the Sierra de Umango ridge a few rem- nants of gravel covered plains more than a hundred meters over the valley bottoms (Quebrada Espinal). — To the east of Villa Union, near to the extreme foot of the Nevado de Famatina there are also remnants of older plains appearing in the form of well marked terraces (Puerto Alegre). Some lower terraces farther out in the lowland are of considerable extensions. They all slope west-
_ ward from the Nevado. In former times they have probably exten-
ded over the greater part of the Pagancillo lowland. The Lomas de Villa Union are but the last withnesses of these plains. Every hill here has a flat summit covered with a gravel sheet (old sheet flood plain).
The best preserved sheet flood cones are, of course, those of younger date (later quaternary?). They, indeed, form almost all lower slopes of the existing mountains. The best developed ones occur on the eastern slope of the Cerro Potrero Viejo (toward the open Vinchina valley). Also the cones on the western side of the Cerro Villa Union (the Guandacol valley) are of wide extension. Giant cones have developed also along the western margin of the Nevado de Famatina.
Recent sheet flood cones seem not to be deveioped, because the last erosion phase, as will be stated below, has only worked by linear way, cutting through the cones and forming canyonlike gorges. The main cones correspond to the main valleys, and the main features of the sculpture seem to have been formed during the last phase of sheet flooding. .
3) A later (recent) erosion phase. Linear cutting. Riversterraees. On studying the great sheet flood cones and their corresponding valleys one finds, that these forms have been to a high degree modified by a erosion, caused by an subrecent uplift of the moun-
tains (or by a relative sinking of the valley floors.) This later phase
128 | SIERRA DE UMANGO
of erosion was thence rejuvenating by way of linear cutting in the sheet flood planes and by! a deepening of the corresponding valley profiles. Some of the cones are to such a degree dissected by the young erosion, as to be changed into a „bad land“ sculpture (north- ern end of the Cerro La Pampa). Many curious hilly forms have been shaped where the sheet flood gravel has endurated to a hard bed, mostly like mesetas and separated by steep canyons. The Lomas of Villa Union, as mentioned, remnants of a great cone- —
SS
<<
ey TEL
IL I ES SS BAND ¢
NS >
>
FIN
N EN - 7 Na
Fig. 23. A wiew of the eastern side of the Sierra de Umango ridge, looking north, along the „inter-
montane zone“. The pic ure demonsrates the formation of piedmont (gravel covered slopes) cut
across by young erosion. The arrow indicates the overthrust plane between the mountain block and the sediment zone. The stream bed of the Quebrada Cordobés in the foreground.
surface extending from the Nevado toward the Pagancillo low- land, have all a table shape, due to the hardened gravel sheet very resistant to the erosion. |
The young erosion has also worked laterally in more advanced stages and thereby reduced the cones into flattopped ridges radia- ting from the apex of the former cone (see the terraces NE from the Villa Union.)
In the valley bottoms, except in the Guandacol valley, there are
EEE: >
GEOMORPHOLOGY 129
in many parts well developed river terraces of quite recent age. These terraces are not of an accumulative, but of a destructive kind, carved out in the sandstone terranes, the substratum of the lowlands.
The most perfect terraces are met with along the lower course of
| Rio Bermejo, between the Villa Union and the Paso de Lamar, that
q
is, in the western part of the Pagancillo lowland. Excellent terraces occur also in the named Paso. It is obvious, that all these terraces
have been formed by a successive throughcutting of the sandstone
ridge of Paso de Lamar, due to a gradual tectonical sinking of the Guandacol valley floor. It may also be assumed, that the breaking
off of the barrier of the Paso de Lamar has adjusted the whole
drainage of the Pagancillo lowland and the connecting Vinchina valley during the last erosion phase. The Lomas of the Villa Union have thence been carved out by this adjustment, as later on the very river terraces along the edge of the flood plain of the Rio Bermejo.
Upward from the Villa Union no more terraces are visible except in its near vicinity to the northeast. The upper part of the Vin- china valley is, as far as I have seen, only an even sandy floor (so called ,medano“), and the sheet flood cones on both sides are not dissected. This part of the valley lies yet aside from the in- fluence of the Paso-de-Lamar-adjustment.
In the Bolson of Jaguel the results of the latest erosion are de-
veloped only to a very small degree. Almost the whole basin is
an flat accumulation plain. The Jaguel river, crossing the plain from NW, has no marked valley, except where it approaches the
sandstone ridge to the east, the outlet-canyon down to the
Vinchina valley. Here a backward erosion carves out the alluvial accumulations of the plain, and some terraces have been formed.
The erosion tends, however, to extend its work over the whole plain
until it would be paralyzed by alluvial depositions from the sur- rounding slopes. The large tectonical depression of Zapallar — Nacimientos —
130 ; SIERRA DE UMANGO
Tambillos, separated from the broader Guandacol valley by the - sandstone ridge of the Cerros de San Antonio (the southern pro- — longation of the „intermontane sandstone zone “), has a very bro- ken bottom, due to a young erosion of old valley floors. This rejuvenation originated from the throughcutting by the Rio La Troya of the named sandstone ridge. Standing on the slope of the mountain, which rises to the east of the Tambillos depression, and looking westward, one obtains an excellent view of the now in- tricately dissected old valley floor, once flattened over the variega- ted series of filling sandstones. The erosion is a subsequent one, following the complicated fault tectonics of the strata. The nume- rous hills carved out in these sandstones show still on their flat- topped summits the remnants of the mentioned ancient valley floor. Upward the slopes the old surface passes into less dissected sheet flood cones. |
The Rio La Troya (which is dry during the greater part of the year) „upstream“ irom the Tambillos depression leeds to the Guan- dacolinos depression, which separates the Cerro Cacho complex from the near situated Cordilleran border. ‘Here, as in the upper
part of the) Vinchina valley, are no signs of young terrace cuts. —
The valley floor is a flat sandy „medano“.
4) Differences of the detailed sculpture in the crystalline rock-ground and in the terranes of | the continental fsediments.
Results of the last erosion phase are visible also in the minute surface forms of the mountain slopes, which dissect the ground in a manner possible only in desert-regions. In the crystalline rock ground such an asperous surface has been formed as to make the forthcoming almost impossible. The only usable trails follow the gravel filled bottoms of the larger quebradas. The numberless sharp ridges have been called in Spanish ,,cuchillas« (= choppers). Above
Å
GEOMORPHOLOGY 131
these slopes nearer to the lofty summits of the mountains the
„euchilla“-sculpture changes into quite a broadlined one (fig. 24).
_ The very summits are often almost plain. This feature is naturally
due to the youthful stage of the erosion, according to which the high parts still remain unaffected by it. |
The „cuchilla“-sculpture of the crystalline rock ground stands in a marked contrast to the detail-forms, developed in the continental sand- stone terranes on the lower hills at the foot of the great blocks and also on the cores of the sheet flood cones, when these have been dissected
Fig. 24. Highland topography near the crest of the Sierra de ano ridge. View to the east with the Nevado de Famatina in the distance. (Height of observation point about 3000 m.)
to a higher degree. The sandstone-forms are generally more rounded and broader than is the case with those in the crystalline ground. The shape of the ridges of the sandstones depends naturally on the dip of the strata. When they lie horizontally or dip gently, the table- or „meseta“ -forms dominate. When they stand vertically the result is a wild Jabyrinth of steep gorges. Such a country is mostly impossible to traverse. Characteristical of the sandstone terranes
are the numerous bluffs or vertical walls, sometimes several hundred
meters in height. They -nay partly be the result of faults, but often they are but erosion forms. The bluffs occur especially in places, where the strata dip under angles of about 30°.
132 | ; SIERRA DE UMANGO
The strata of the quaternary and recent accumulations in the j
valleys being much softer are sometimes dissected into „bad land"- like forms, but generally they are not affected by erosion.
At last may be communicated a scheme of the principal sculpture- zones as they depend on level and nature of the mother-rock.
Flat topped summits and undulating upland surfaces.
i Jagged (,cuchilla*) sculpture, draining to the Old crystalline quebradas.
rocks.
High destructive terraces. Continental i sandstones cones or sculpture with i bluffs and rounded forms. Gorges.
,Bad land" sculpture.
River terraces, plains.
Quaternary and recent accumulations.
Continental sandstones. Soft accumulation masses. Alluvium.
H) MINERAL OCCURRENCES.
Selenide bearing veins.
In comparison with the Nevado de Famatina and its famous rich- ness of metalliferous minerals, well known also in the Anglo-Ame- rican mining world, the neighbouring mountains to the west, the Sierra de Umango region, have but insignificant occurrences of exploitable minerals to offer. The most worked ones are the silver- selenide bearing veins in the Cerro Cacho complex. The first dis- covery of these was made by the mining engineer E. Htinicken,
engaged in the Famatina Copper Company. Later on professor L. :
Brackebusch collected several specimens of these selenides du- ting his expeditions in the years 1875--1888. The specimens of Brackebusch were then more closely examined in Germany.!
1 Otto: Eukairit aus Argentinien. Berichte der deutschen Chemischen Gesell- schaft. Jahrgang XXIII. 1890.
F. Klockmann: Ueber einige seltene argentinische Mineralien. Zeitschr. für Krystallographie. Band XIX. 'N:o 3. Leipzig 1891.
MINERAL OCCURRENCES 133
The chief selenide veins lie on the eastern slope of the Cerro Cacho, some leagues to the west of the Puesto Umango. The veins have been exploited some ten years ago by an Argentine mining - prospector. The claims are still visible and well accessible, but all valuable parts of the mineral mass have been quarried. Without operating with dynamite it seems impossible to obtain any good specimens of the minerals. The selenides, eukairite and umangite, have been exploited for their high value in silver and copper. The _ ore was brought on muleback over the Sierra de Umango ridge and the Paso de Tocino over the Sierra de Famatina down to the town of Chilecito along asperous and even dangerous trails. The mines lie very high, about 3000 m and the icecold storm from the Cordillera sweeps nearly the year round over these heights.
The selenide minerals mentioned occur in thin fissure veins transversally cutting across the crystalline schists. These last named consist principally of limestone and amphibolite, contorted in many ‘ways. The gangue mass is for the most part calcite and quartz. The metalliferous minerals occur as scattered grains or irregularily formed dense masses. According to Klockmann (I. c.) these mi- nerals are, as already stated, eukairite and umangite. The former mineral has the composition AgCuSe, the later Cu,Se,. The eukai- tite is as specimen known already many years ago, principally from Småland, Sweden.! The Umango-eukairite was, as said, first found and examined by Hiinicken. The umangite was, however, fixed by this explorer as bornite, but the true nature was then stated by Klockmann. The chemical composition of the two minerals is according to him as follows:
Eukairite: Umangite: N na er ADs OS SD ER RE 54 Yo Sie ea tile DAs eee AAS) chr ak pes ae oat OR a bref del Aes SIR N et RE SOE PR AOL a RRL tach all ou traces.
1A. E. Nordenskiöld: Ofversigt af Kongliga Vetenskapsakademiens Hand- lingar. Stockholm 1886.
134 | SIERRA DE UMANGO
Among the mineralogical characteristics of the two species may be mentioned, that the eukairite has a lead-white, granular habit with a brilliant lustre. The umangite shows in fresh fractures a
reddish, dark colour and a high lustre. The habit is massive. |
Crystals of both minerals are unknown.
Other minerals associated with these are bornite and of secondary products principally malachite.
The veins have no visible relations to eruptives. They are con- siderably younger than the granites.
In the Umango area there are also some other occurrences of selenides, but of less importance. On the trail between the Puesto Umango and Tambillos there lies in a very broken ground a small mine called „la Millionaria“. The chief rock here is a mica schist and the veins are as in the former case transversal. Malachite alte- ration products are also here very abundant.
In the Quebrada Agua de los Caballos situated in the eastern part of the Cerro Potrero Viejo another transversal vein cut through a gneissic rock. It contains abundantly hematite, gathered along the salbands of the vein. This occurrence was recently exploited (by a French prospector, 1914 gone off to the war).
Bodenbender (1911) mentions the occurrence of umangite- and clausthalite(PbSe)-bearing veins with a calcite gangue mass, cutting through the Famatina granite in the vicinity of the Piedra Pintada in the Sierra de Safiogasta. Outside the Umango area there are selenide bearing veins to the north in the province of Catamarca, where umangite has been found in the mountains to the west of Tinogasta. No further facts are known about the ap- pearance of the veins here.! | ;
Other selenide occurrences in the Argentine Republic are met with in the province of Mendoza, in the Cerro Cacheuta mining camp.
1 G. Bodenbender: Communicaciones mineras & mineralögicas. VIII. Cria-
deros de seleniuros de cobre etc. Boletin de la Academia Nacional de Ciencias en Cordoba. Tomo XVII. Cordoba 1903.
MINERAL OCCURRENCES 135
Since the discovery by Domeyko! a large series of papers” has already been published on these mineral findings. The veins set up in a porphyritic rock, which according to a verbal communication by dr Stappenbeck is of lower Gondwana age. The porphyrite sheet is crossed by numerous eruptive veins of young date, and these veins may be supposed to stand in genetic connec- tion with the selenide occurrences. The selenide minerals here have a very varying composition and all the specimens have different names (see Domeyko, 1. c.).
It seems probable, that the Sierra de Umango and the other oc- currences of selenide minerals belong to the andine volcanic phase during which all the famous ores of the Cordillera and the Nevado de Famatina were formed. The straight cutting through of the Umango veins is also indicative of a very young age.
Occurrences of nickel ores.
Mineral occurrences of economic value, except the described sele- nide veins, have been found north of the Umango mountains, but in the same geological structure, where it reappears at the northern border of the alluvial plain of the Jagucl bolson. According to Brackebusch? there lies near to Potrero Grande a mine called „Solitaria“ and containing niccolite (NiAs) with 44 9/0 Ni, the only finding of this mineral in the Argentine. Annabergite |(AsO,);Ni, + 8H, O] has been encountered also. The veins here cross an am- phibolite, associated with quartzitic and gneissic rocks. The mine was exploited during 1845 by two Germans, the Erdmann brothers. But the many political complications at that time in Argentine made it impossible for the miners to carry on their work.
1 A. Domeyko: Anales de la Universidad Nacional de Santiago de Chile.
Tomo XXIX. 1876. Pages 62-—68.
2 See Stelzner (1876—85). Page 219. 3 Ludw. Brackebusch: Die Bergwerkverhältnisse der argentinischen Repu- blik. Zeitschr. für Berg-, Hütten- und Salinenwesen im Preussischen Staate. Bd.
E41. 1893.
BIBLIOGRAPHY.
(More closely referring to the Sierra de Umango-geology)
1876 — B. H. Geinitz: Ueber rhaetische Pflanzen- und Thierreste in den
1876 — E. 1876—85 — 1891 —
1891 —
1891 —
1893 — 1894 — 1894 I. 1895 —
1896 —
1896 —
L.
B.
argentinischen Provinzen La Rioja, San Juan und Mendoza. Berlin und. Cassel. | Kayser: Ueber primordiale und untersilurische Fossilien aus der argentinischen Republik. Berlin u. Cassel.
A. Stelzner: Beiträge zur Geologie und Paläontologie der argen- tinischen Republik. I Teil. Berlin und Cassel. (II Teil, see above.) Brachebusch: Mapa geolögico del interior de la Reptib:ica Ar- gentina. Escala 1: 1,000,000. Gotha.
Kühn: Untersuchungen an altkrystallinischen Schiefergesteinen aus dem Gebiete der argentinischen Republik. Neues Jahrbuch für Mineralogie, Geol. und Pal. Beilageband VII. Stuttgart.
. Romberg: Petrographische Untersuchungen an argentinischen
Graniten. Ibidem. VIII.
Brackebusch: Ueber die Bodenverhältnisse des nordwest- lichen Teiles der argentinischen Republik. Petermanns Mitteilungen. Band XXXIX. Page 153—166.
Paul Siepert: Petrographische Untersuchungen an alten Erguss-
gesteinen aus der argentinischen Republik. Neues Jahrb. für Min. etc. Beil. Bd. IX. Stuttgart.
Bodenbender: Devono y Gondwäna en la Repüblica Argentina. Boletin de la Academia Nacional de Ciencias en Cérdoba. Tomo XV. Cördoba.
K u r t z: Sobre la existencia del Gondwana inferior en la Repüblica Argentina. Revista del Museo Nacional de La Plata. Tomo VI. La Plata.
. Bodenbender: Sobre la edad de algunas formaciönes carboni-
feras en la Republica Argentina. Revista del Museo de La Plata. Tomo VII. La Plata.
Beobachtungen iiber Devon- und Gondwanaschichten in der argentinischen Republik. Zeitschrift der Deutschen Geolo- gischen Gesellschaft. Band 48. Heft 4 Berlin.
1897 — — — — —: Devono y Gondwana en la Republica Argentina. Bo-
letin de la Acad. Nacional de Ciencias en Cördoba. Tomo XV. Entrega 2—3. (ördoba.
138
BIBLIOGRAPHY
1902 — —— ————: Contribuciön al conocimiento de la Precordillera de
1911 — G.
1913 — W.
1914 — H.
1914 — O.
1916 — J.
1916 — J.
San Juan y Mendoza y de las Sierras centrales de la Republica Argen- tina. Boletin dela Academia Nacional de Ciencias en Cérdoba. Tomo XVII. Cordoba.
—: Contribuciön al conocimiento geolögico de la Republica Argentina. I. Petrografia. Anales del Ministerio de Agricultura de la Naciön. Secciön Geologia. Tomo II. N:o 3. Buenos Aires.
. Stappenbeck: La Precordillera de San Juan y Mendoza. Ana-
les del Ministerio de Agricultura de la Naciön. Secciön Geologia. Tomo IV. N:o 3. Buenos Aires.
. Keidel: Die neueren Ergebnisse der staatlichen geologischen Unter-
suchungen in Argentinien. Compte Rendu. Congrés Géologique International. XI Session. Stockholm. Bodenbender: Constitucién geolögica de la parte meridional de la Rioja y regiones limitrofes. Boletin de la Academia Nacional de Ciencias en Cordoba. Tomo XIX. Cordoba. (With a geologi- cal map).
Penck: Mapa geolögico de la parte occidental de la Provincia de Catamarca (Valle de Fiambala and surroundings). Original in the Museum of the Direction of Mines. Buenos Aires. Scale 1 : 200,000.
La geologia del valle de Fiambalä y de las Cordilleras
limitrofes al norte de Tinogasta (Provincia de Catamarca). Memo- ria de la Direccién General de Minas. Ministerio de Agricultura de la Naciön. Tomo XI. N:o 4. Febr. Buenos Aires. Keidel: Ueber das Alter, die Verbreitung und die gegenseitigen Beziehungen der verschiedenen tektonischen Strukturen in den argen- tinischen Gebirgen. Compte Rendu. Congres Géologique Interna- tional. XII Session. Ottawa. Stiglitz: Contribuciön a la petrografia de la Precordillera y del Pie de Palo. Ministerio de Agricultura de la Naciön. Direcciön Ge- neral de Minas etc. Boletin X. Serie B. Buenos Aires. |
(Cordillera NW-Argentiniens). Neues Jahrbuch fär Mineralogie etc. Beil. Bd. XXXVIII. Stuttgart.
Rassmuss: Rasgos geolögicos generales de las sierras pampea- nas. Ministerio de Agricultura de la Naciön. Direccién Gral de Minas, etc.. Boletin XIII. Serie B. Buenos Aires.
. Bodenbender: EI Nevado de Famatina. Boletin de la Aca-
demia Nacional de Ciencias en Cördoba. Tomo XXI. Pag. 100—182. (Map only in original).
Rassmuss: Geological maps: 1) Sierra del Aconquija, provincia de Catamarca. 2) La Precordillera Riojana. Both in scale 1 : 200,000. Originals.in the Direction of Mines, Buenos Aires.
.Penck. Der Südrand der Puna de Atacama. (NW-Argentinien). Abhandl. d. mat.-phys. Klasse der sächsischen Akademie der Wissenschaften. Bd. XXXVII. N:o 1. Leipzig. With geological and topographical maps in the scale 1 : 200,000.
5 kas SoS JAR SA
. Penck: Hauptzüge im Bau des Südrandes der Puna de Atacama -
Explanatory note to the maps.
The topographical map was worked out during the wintermonths 1915 by the topographer of the Direction of Mines, Buenos Aires, mr Jorge Schultz contemporarily with my own field works. The original field sheets in greater scale were reduced to the scale 1:100,000 and connected to form a map of the whole Sierra de Umango area, except the western and northern parts of the Cerro Cacho complex, which remained unmapped. The present map shows also nothing of the mountains more to the west, i. e. the very border of the Cordillera, which the Sierra de Umango com- plex joins (Pampa de Leoncito and surroundings). Neither the wes- tern side of the Guandacol valley — the Precordillera Riojana — nor the eastern wall of the Vinchina valley — the Nevado de Fa- matina — are shown on the map. These mountains have, however, been mapped in other connections. To get a better orientation of the main topographic features in this part of the Republic a map of smaller scale should be” used. The best existing is the hypso- metric map of Ludw. Brackejbusch (1893).
The geological map reproduced on the cover-sheet was pre- pared after the field work were finished and as soon as the topo- graphical sheet was to the writers disposal. Thence only the main geological features were fixed, partly because the scale of the map is reduced to a considerable degree. I believe, however, that the quite numerous figures in the texte may be a kind of une ee ‚to the geological picture of the area in question.
is
Pea
"
PIAGET.
Fig. 1. A recrystallized mica — quartz schist. The dark, parallel stripes are biotite, the large grains quartz. Nic. +. Magn. 28 X. Cerro Potrero Viejo. Quebrada Agua de los Caballos.
/
Fig. 2. Sericitic, microfolded ferriferous schist. Nic. +. Magn. 20 X. Sierra de Umango ridge, eastern side. Quebrada Salto (N from Casa Piedra).
PEATESE:
Fig. 1. Famatina granite (aplitic type) of primary habit. The pale individuals are quartz, the darker perthitic potash feldspar. Nic. +. Magn. 20 X, Sierra de Sano- gasta. On the road Puerto Alegre—Sanogasta.
Fig. 2. Famatina granite, crushed. Nic. +. Magn. 20 X. Sierra de Sanogasta. On the road Puerto Alegre-—-Sanogasta.
PLATE If,
Fig. 1. Amphibolite of granoblastic, recrystallized structure. From a breccia frag- ment in granite. Nic. 7. Magn. 88 X. Cerro La Vatea, opposite Villa Castelli.
Fig. 2. Biotite transformed to fibrolite (sillimanite). Syntectic rock, surrounded by granite. Nic. +. Magn. 88 X. Cerro La Vatea.
PLATE IV.
Fig. 1. Amphibolite schist, to some degree metamorphosed by the white granite. The amphibole shows a beginning alteration to biotite, due to the acidification of the rock. The specimen was taken from a transition band between amphibolite and granite. The pale component is plagioclase and also quartz, the darker are amphibole and biotite. Nic. 7. Magn. 20 X. Quebrada Agua de los Caballos, Cerro Potrero Viejo.
Fig. 2. Dark quartzite rock from an inclusion in young granite. Pavement horn- fels structure. The black spots are biotite. Nic. 7. Magn. 120 X. Steep rock hill rising from the valley floor to the west from Nacimientos.
PLATE V.
Fig. 1. A dynamic flow structure developed in a white aplitic granite, which forms a concordant vein in a mica schist. Nic. 7. Magn. 20 X. Western foot of the Cerro Villa Union, Carrizal.
Fig. 2. Red aplite granite, completely granulated. Nic. 7. Magn. 28 X. The western foot of the high ridge of the Sierra de Umango at Arroyo Gaucho.
EIERN Le N ne
FRA HÄ RED
Bi MDa ER bel
GAY 5 or
TA Seer } ante i
O AREA
esent copy 1:400,00(
fixed by H. Hausen. ST. o mu =
Lowland alluvium, /\ NS ‚Piedmont gravel fans.
Ra One 0 0% 90) | ternary and tertiary).
Conglomerates, coarse, dislocated (qua- )
ntinental, detrilic formations (ter jary - mesozoic upper palaeozoic).
e- intruded metamorphic schists crystalline basement).
lepmontane eS — 4. Cerro Bolten NER (Cerro
r
NA
Rue:
M
ÄN
KEN IK
ct VER a VN Dunn Ze A lie u. ANNE ha ‘HE SIERRA DE UMANGO AREA. year 1915 by the Topographer of the Direction of Mines, Buenos Aires, Mr Jorge Schultz,
p se ni
“Seale of the original map 1:100,000, Scale of the present copy 1:400,000,
ey ( hive r 4 SAN TR SL FÅ 1: ips
ACTA ACADEMIAE ABOENSIS MATHEMATICA ET PHYSICA I:5
DIE LICHTEMISSION UNTER DER WIR- KUNG MOLELEKTRISCHER KRÄFTE AN DER OBERFLÄCHE VON KRISTALLEN
VON
Dr. BRUNO SAXEN
LEKTOR DER PHYSIK A. D. AKADEMIE ÄBO (FINLAND).
ÄBO AKADEMI ÄBO 1921
Se =
= ER
INHALT.
SEITE SW EENDBERUNDEGRUNDGEDANKEN 20.2 ei tied a ee 5 AU SHU MROUNG DER IME DHODE ice ee ae be a) Oe J Been 7 EMISSION EINES KONTINUIERLICHEN SPEKTRUMS DURCH KANAL- SHEP SAN OE BN Sai le ae ER a RMS, rs ane ro fee RC ey ey AY RE 10 AUFTRETEN VON KANALSTRAHLENLINIEN IM LICHT DER OBER- BIRNSRIEINSCHICHIT AR. ET a ee ee 11 EINFLUSS MOLELEKTRISCHER FELDER AUF DIE LINIEN VON KA- DESSEN (BI BINS Dre ARS eo ee Eg a ar ar 12 EINFLUSS MOLELEKTRISCHER FELDER AUF DIE LINIEN VON AB- GETRENNTEN!METAELATOMEN: (Naj Ca) an ae. 12
ERGEBNISSE 2. tic eke ale sheet len ie SE SIG EA Nats /d ashe ig ale RAR AG 15
DIE LICHTEMISSION UNTER DER WIRKUNG MOL- HJÖEKTRISCHER ‚KRÄFTE AN DER OBERFLACHE VON KRISTALLEN.
1. AUFGABE UND GRUNDGEDANKE. Es ist zu vermuten, dass an der Oberfläche von festen Körpern, so von Kristallen, elek- trische Felder aus der Oberfläche herausspringen, die beispiels- weise bei katalytischen Reaktionen wirksam sind, wie Professor Haber vermutet. Stark! hat es für möglich gehalten, dass das Auftreten solcher molelektrischen Felder auch einen Einfluss auf Spektrallinien, die unmittelbar an der Oberfläche emittiert werden, ausüben könnte. Man kann vermuten, dass der Durchmesser dieser Felder mit dem der molekularen Wirkungssphäre (10° cm) über- einstimmt. Wenn darum solche Felder vorhanden sind, ist zu erwarten, dass elektrisch empfindliche Serienlinien (grosser Stark-Effekt) ver- breitert werden, so weit ihre Emission aus einer unmittelbar an der Oberfläche liegenden (10 °cm dicken) Schicht kommt. Andererseits müssten elektrisch unempfindliche Linien (kleiner Stark-Effekt), die aus einer solchen Schicht kommen, keine Verbreiterung zeigen. Zur Prüfung dieses Gedankens ist eine Methode zu finden. Erstens muss an der festen Oberfläche eine sehr intensive Lichtemission ‚angeregt werden. Zweitens muss: dafür gesorgt werden, dass aus- schliesslich aus einer sehr kleinen Schichtdicke Licht zur Beobachtung kommt, dass also Licht aus grösserer Entfernung von der Ober- fläche ausgeschaltet wird. Nach einem Vorschlage von Professor Stark lässt sich diese Aufgabe folgendermassen lösen. Man ver-
1 J. Stark, Nobelvortrag: Änderungen der Struktur und des Spektrums che- mischer Atome. Hirzel, Leipzig 1920. Seite 11.
6 LICHTEMISSION A. D. OBERFLÄCHE V. KRISTALLEN
wendet zur Lichterregung Kanalstrahlen, die auf eine feste Ober- fläche fallen. Sie rufen zwei Wirkungen hervor: erstens spalten sie 2 Atome von der Oberfläche ab; zweitens erregen sie die abgespal- tenen Atome zur Lichtemission an. Diese leuchten am stärksten un- mittelbar an der Oberfläche; die Lichtintensität nimmt mit der Entfer- nung von dieser ab. Um das Licht nur aus der unmittelbar an der Ober-. fläche liegenden Schicht zu erhalten, verfährt man folgendermassen:
Man verwendet einen festen durchsichtigen Körper, lässt Kanal- strahlen unter 45° auf die gut polierte Fläche fallen und beobachtet senkrecht zur Kanalstrahlenrichtung die Lichtemission der Ober- fläche (Fig. 1). Aus nebenstehender Zeichnung ist zu ersehen, dass Licht nur aus dem Kristall oder aus der Schicht unmittelbar an seiner Oberfläche
-]DUDyY
kommen kann; Licht von grösserem Ab- stand kann nicht austreten. Ferner wird
<——__ vayynajs
eine Schicht, deren Dicke ungefahr gleich
einer Wellenlänge des Lichtes ist, hinsicht-.
lich der Emission der elektromagnetischen
— isang Lichtwellen an der festen Oberfläche teils
zum festen Körper, teils zum Gasraum zu
Ei rechnen sein, weil eine ebene Fläche sich
erst in einem Abstande, der grösser als
eine Wellenlänge ist, sich definieren lässt, da ja die linearen Ab- messungen der Unebenheiten von der Grössenordnung einer Wel- lenlänge sind. Bildet man also die Auftrefistelle der Kanalstrahlen auf der Hypotenusenfläche mit einem Objektiv, dessen Achse senk- recht zur Kathetenfläche ist, bei kleiner Öffnung ab, so ist ım Bilde Licht nur vom festen Körper bezw. von der Schicht unmittel- bar an der Oberfläche vorhanden; das Licht aus grösserem Abstande erscheint nicht im Bilde und kann also nicht störend einwirken. Durch okuläres Anvisieren konnte dies bestätigt werden. Im Leucht- fleck an der Oberfläche waren leicht zweierlei verschiedene Leuchten
zu unterscheiden, wenn man das Auge vor der Kathetenfläche des
AUFGABE UND GRUNDGEDANKE 7
Kristalls bewegte. Das eine Leuchten erwies sich unabhängig von der Bewegung des Auges, hatte also seinen Ursprung an der Ober- fläche selbst, während das andere seine Stelle veränderte, musste also vom Gasraum an der Oberfläche herrühren. Die zwei Emissions- gebiete unterschieden sich auch von einander in ihrer Farbe; das vom Kristall herrührende Licht war z. B. bei Sauerstoffkanalstrahlen auf einem Flusspatkristail bläulich, das Licht von der Schicht vor der Oberfläche rötlich. !
2. AUSFÜHRUNG DER METHODE. Bei den Untersuchungen wurde eine Entladungsröhre von dem aus Fig. 2 ersichtlichen Bau verwen- det. Der Durchmesser der Röhre be- trug 4,5 cm, der Abstand zwischen ?"”?? Wir den Aluminiumelektroden 19 cm und die Öffnung in der Kathode 1,; mm.
Diese Öffnung konnte mittels einer an Neger kleinen, an der Kathodenfläche dreh- os bar befestigten Metallscheibe von aus- sen mit Hilfe eines Magneten ge- schlossen werden, sodass die Kanal-
— Gas
strahlen, solange sie noch unrein Fe
waren, verhindert wurden auf die Kristallfläche zu fallen und diesel- be mit fremden Stoffen, besonders Kohlenstoff, zu überziehen. Diese Schutzanordnung erwies sich sehr notwendig, besonders bei Was- serstoffkanalstrahlen, denn die grossen Mengen Kohlenstoff, die im Anfang in der Entladungsröhre vorhanden waren, verursachten, dass die Kristallflache, wenn die Kathodenöfinung nicht zugeschlossen war, sehr bald mit Kohlenstoff überzogen wurde, wodurch das im Kristall erscheinende Licht in kurzer Zeit abgeschwächt wurde. Die kleinen Spuren von Kohlenstoff, die ungeachtet vielstündigen
1 Eine ähnliche Beobachtung hat schon Goldstein gemacht. Siehe: Hand- buch der Radiologie, Bd IV, S. 5.
8 LICHTEMISSION A. D. OBERFLÄCHE V. KRISTALLEN
EN fe
Durchspülens mit Wasserstoff unter gleichzeitiger starker Entladungs- |
stromstärke immer noch nicht ganz entfernt werden konnten, verur- sachten bei den ersten Versuchen viel Verdruss. Wegen der kleinen Lichtintensität im Leuchtfleck musste das Exponieren auf mehrere
Stunden ausgedehnt werden, was wieder erfolglos blieb wegen der Ab-
schwächung des Leuchtens mit der Zeit durch die auf der Kristall- fläche niedergeschlagene Kohlenstoffschicht. So konnte ich gar keine verwendbare Auinahme von Wasserstofikanalstrahlenlicht auf Fluss- patkristall erhalten. Erst als ich mit Sauerstoff die Röhre längere Zeit unter gleichzeitiger Entladung durchspülte, wurde der Kohlen- stoff in der Röhre weggebrannt und die Kristallflache blieb sauber während des ganzen Versuches. Erst nachdem die Röhre durch Sauerstoffentladungen gereinigt worden war, gelang es mir eine sehr gute Aufnahme von Wasserstoffkanalstrahlenlicht auf Steinsalz zu bekommen.
Unmittelbar hinter der Kathode waren die Kristalle auigestellt. Die Lichtemission wurde an Flusspat und Steinsalz beobachtet. Die Kristalle waren in der Form eines Prismas geschliffen, dessen Durchschnitt ein gleichschenkliges rechtwinkliges Dreieck bildete. Die Länge der Kathete betrug beim Flusspat 2 cm, beim Steinsalz 1,5 cm. Um störendes Licht auszuschalten, wurden die Kristalle in dünnes Aluminiumblech eingehüllt. Auf der Hypotenusenfläche war im Metallblech eine kreisförmige Öffnung von etwa 5 mm Durchmesser ausgeschnitten, so dass die Kristalliläche frei blieb, um die Kanalstrahlen auizufangen. In derselben Höhe war im Blech auf der vorderen Kathetenfläche eine ähnliche Öffnung angebracht,
durch welche das Licht beobachtet werden konnte. Die Kristalle :
waren so auigestellt, dass sie in zwei zu einander senkrechten Richtungen verschoben werden konnten, wodurch das Kanalstrah-
lenbündel sich leicht auf die freie Kristallflache einstellen liess.
Gleichachsig mit dem gegenüber der Kathetenfläche des Kristalles angebrachten Seitenrohr, das mit einer Glas- bezw. Quarzplatte verschlossen war, stand das Spaltrohr des Spektrographen. Im
AUSFÜHRUNG DER METHODE 9
sichtbaren Spektrum verwendete ich einen lichtstarken Zeiss-Tessar- ' Spektrographen (Öffnungsverhältnis 1:3,5), im Uitraviolett machte ich die Aufnahmen mit einem Steinheil-Quarz-Spektrographen (Öffnungsverhältnis 1:14). Im ersten Falle wurde der Leuchtfleck an der Kristallflache mit einem Zeiss-Tessar-Objektiv scharf auf dem Spalt des Spektrographen im Verhältnis 1:1 abgebildet, im zweiten Falle mit einem Flusspatachromat.
Die Aufnahmen wurden auf Hauffs Ultrarapid- und Flavin-Platten genommen- und die Exposition auf 5—12 Stunden ausgedehnt. Gleich nach Beendigung der Exposition wurde ein Vergleichsspektro- gramm unter ganz ähnlichen Verhältnissen mit Helium- oder Queck- silberlicht genommen. Die Platten wurden mit Hauffs Rodinal entwickelt. Mit einem Hartmannschen Mikrophotometer wurden die Linienabstände und die Schwärzungsverhältnisse bestimmt.
Als Stromquelle wurde hauptsächlich eine mehrplattige Iniluenz- maschine verwendet. Meine Absicht war zwar, mit einer Hochspan- nungsdynamomaschine (3,500 Volt) zu arbeiten, aber die Spannung erwies sich zu klein, um ein zureichend intensives Leuchten an der Kristallflache zu erzeugen. Alle Aufnahmen, die mit niedriger Span- nung gemacht wurden, waren zu sehr unterexponiert, um aus ihnen einige Schlussfolgerungen zu ermöglichen. Erst bei einer Spannung von über etwa 5,000 Volt erschien der Leuchtfleck genügend intensiv.
Die Versuche wurden, wie schon bemerkt, mit Wasserstoii- und Sauerstoffkanalstrahlen gemacht. Durch Erhitzen eines im Entla- dungsrohr eingeschmolzenen Palladiumröhrchens mit einer Wasser- stofflamme konnte das Wasserstoffgas in das Rohr hineingelassen werden, während die Gaede-pumpe gleichzeitig arbeitete. Durch Regulieren der Grösse der Flamme und der Umlaufsgeschwindigkeit der Pumpe konnte während des ganzen Versuches ein dauernder Gasstrom im Entladungsrohr unterhalten werden. Ein gleichmässi- ger Gasstrom war auch bei Sauerstoff, der durch elektrisches Er- hitzen von Kaliumpermanganat dargestellt wurde, leicht durch An- passen der Stärke des Hitzstromes zu erhalten.
10 LICHTEMISSION A. D. OBERFLÄCHE V. KRISTALLEN
Nach jedem Versuch wurde der Kristall aus der Röhre heraus- genommen und, wenn nötig, gereinigt. An der Auftreffstelle der Kanalstrahlen war eine deutliche Vertiefung zu beobachten, die ganz glatt und durchsichtig war.!
3. EMISSION EINES KONTINUIERLICHEN SPEKTRUMS DURCH KANALSTRAHLEN.. Mit der Erregung der Lichtemission durch den Stoss der Kanalstrahlen haben sich mehrere Forscher beschäf- tigt, und dabei ist u. a. festgestellt worden, dass das an festen Me- tallverbindungen, Salzen und Oxyden erregte Licht mehr oder min- der grosse Bezirke des Spektrums kontinuierlich ausfiillt.2 J. Stark und G. Wendt haben in einer Arbeit? die Frage nach dem Orte dieser Bandenemission und nach ihrem Zusammenhang mit dem Stoss der Kanalstrahlen zu beantworten versucht. Dabei sind sie zu der Schlussfolgerung gekommen, dass der Ort der Bandenemis- sion die Gesammtheit der Auftreffstellen von Kanalstrahlenteilchen ist. Dieses direkt angeregte Licht wird dann bei seiner Ausbreitung in den festen Körper hinein in Bandenfrequenzen von Atomen des Körpers absorbiert und in diesen Banden wieder als Fluoreszenz- licht ausgestrahlt.
Auf allen meinen Aufnahmen, sowohl bei Flusspat wie bei Stein- salz, erscheint das kontinuierliche Spektrum sehr intensiv und er- streckt sich von blau bis etwa 280 uu. Die Intensität nimmt nach der ultravioletten Seite stark zu. Vermutlich geht das Spektrum weit in das Ultraviolett hinein. Dass der Ort dieser kontinuierlichen Lichtemission in der Oberfläche des Kristalles zu suchen ist, ist durch die Versuchsanordnung sichergestellt; die oben angeführte, von Stark und Wendt gezogene Schlussfolgerung in dieser Be- ziehung ist also durch meine Versuche bestätigt worden. Der grösste
1 Vergl. J. Stark und G. Wendt: Uber das Eindringen von Kanalstrahlen in feste Körper. Ann. d. Phys. 38..1912. S. 921.
2 J. Stark: Die elementare Strahlung, S. 230.
3 Ann. d. Phys. 38. 1912. S. 690.
KANALSTRAHLENLINIEN 11
Teil des von Kanalstrahlen erregten Lichtes ist wahrscheinlich Licht der kontinuierlichen Emission und dürfte übereinstimmen mit dem von Kathodenstrahlen, angeregten. In der Oberfläche werden die hohen Frequenzen am stärksten angeregt und möglicherweise wird ein Teil von diesen durch Absorption in unmittelbarer Nähe der Auftreffstelle in niedrigere Frequenzen verwandelt. In meinen Ver- suchen ist die Intensität von grösserer Tiefe unmerklich klein. Durch die ungeordneten Stösse der Kanalstrahlenteilchen auf die
Atome der Oberfläche, wobei die Valenzelektronen mehr oder minder
aus ihren Gleichgewichtslagen abgelenkt werden und bei der Rück- kehr in dieselben verschiedene Frequenzen ausstrahlen können, ist die Emission eines kontinuierlichen Spektrums verständlich. Das Nachleuchten der Auftreffstelle der Kanalstrahlen während mehreren Sekunden nach Abbrechen der Entladung, was immer deutlich beob- achtet werden konnte, knüpft sich wohl an Schwingungen von .all- mählich in ihre Gleichgewichtslagen zurückkehrenden Elektronen.
4. AUFTRETEN VON KANALSTRAHLENLINIEN IM LICHT DER OBERFLÄCHENSCHICHT. Ausser dem kontinuierlichen Licht treten die Linien der Metalle im Kristall, Natrium und Calcium, auf. Diese stammen aus der Schicht unmittelbar an der Oberfläche. Durch diese Schicht dringen die Kanalstrahlen, bevor sie die Ober- fläche treffen. Man darf also erwarten, dass ausser den Linien von Ca und Na, auch die Linien der Kanalstrahlen selber auftreten. Dies be- stätigt sich in meiner Aufnahme an Steinsalz (Wasserstoff-Kohlen- stoff-Kanalstrahlen). Es treten nämlich die Kohlenstofflinie A=4270 A, ferner die Linien Hg und H, auf. Dagegen fehlen in meiner Auf- nahme mit Flusspat und Sauerstoffkanalstrahlen die Sauerstofflinien. Dieses Fehlen der O-Linien ist kaum aus der geringen Intensität zu erklären, denn auf derselben Aufnahme treten sie im Kanalstrahlen- spektrum, das gleichzeitig durch besondere Einstellung erhalten wurde, auf. Vielmehr rührt es vom Erlöschen der Lichtemission der Sauerstoffkanalstrahlen in der Schicht an der Kristallfläche her, wo
12 LICHTEMISSION A. D. OBERFLÄCHE Vv. KRISTALLEN :
zahlreiche, aus dem Kristall austretende Atome vorhanden sind. Eine ähnliche Erscheinung hat Wilsar! beobachtet, nämlich dass Wasserstoffkanalstrahlen, in Quecksilberdampf hineingeschickt, zu leuchten aufhören; sie regen aber die Hg-Atome zum Leuchten an. Das verschiedene Verhalten der Wasserstoff- und der Sauerstofikanal- strahlen in der Oberflächenschicht des Kristalles dürfte im elek- trischen Charakter des Trägers zu suchen sein. Das Fehlen der Sauerstofilinien im Spektrogramm ist andererseits ein Beweis dafür, dass Licht aus grösserem Abstand senkrecht zur Kathetenfläche nicht austritt, sondern nur von der von Kanalstrahlen getroffenen Fläche.
5. EINFLUSS MOLELEKTRISCHER FELDER AUF DIE LINIEN
VON KANALSTRAHLEN. Wenn die molelektrischen Felder an der. Oberfläche eine Wirkung auf die aus dem Kristall austretenden
Metallatome hervorbringen, so muss dieser Einfluss auch an den Kanalstrablenlinien hervortreten. Es werden also im Spektrum des Leuchtfleckes elektrisch empfindliche Linien verbreitert erschei- nen, elektrisch unempfindliche dagegen scharf. Dieses ist in der Tat der Fall. Die Kohlenstofflinie, a=4270 A, ist scharf. Nach Mit- teilung von Prof. Stark konnte er an dieser Linie, selbst in einem Feld von 100,000 Volt/cm, bei mässiger Dispersion, keine Einwirkung feststellen. Dagegen erscheint die Wasserstofflinie Hz stark ver- breitert, ohne definierte Ränder. Die Lage der Linie Hy lässt sich
nur durch Auftreten eines Intensitätsmaximums feststellen; ohne er-
kennbare Abgrenzung geht sie in kontinuierlichen Grund über.
6. EINFLUSS MOLELEKTRISCHER FELDER AUF DIE LINIEN VON ABGETRENNTEN METALLATOMEN (Na, Ca). Wie bereits aus früheren Arbeiten? bekannt ist, treten im Leuchtflecke die Linien des Metalls, also in meinen Versuchen diejenigen von Ca und Na, aul. In der Tabelle 1 sind die in meinen Spektrogrammen auftretenden Calciumlinien zusammengestellt. |
1 Dissert. Würzburg 1912 und Ann. d. Phys. 39. 1912. S. 1251. 2 Stark und Wendt: Ann. d. Phys., 38. 1912. S. 690.
| |
EINFLUSS MOLELEKTRISCHER FELDER 13
Tabelle 1, Ca-LINIEN. Bezeichnung | Lee Intensitat | N Aussehen 4586 mässig 25 Scharf f 4457 | Triplet 4435 |( gross 23 : EN ae N mässig 22 = 4303 gross 4299 | 4989 ziemlich 20 n 4283 | gross 4227 sehr gross 18 | 3 Duplet. f|- 3968 \ f Nach rot LS. \ 3933 J Se lit B10? My | verbreitert Duplet f 3737 I f Nach rot St \ 3706 J Ei 13 | verwaschen | | 3644 | Geringe Triplet | 3630 klein 14 | Unschärfe IN.S. 3624 nach violett N | 3362 | | Unscharf, Triplet | 3350 | sehr klein 11 | ganz ver- INS. 3345 waschen Duplet N 3179 \ { Erste Linie nach TOSS 9 3159 J 5 | rot verstärkt
FS
Nach Untersuchungen von Stark und Kirschbaum! sind die Calcium-Linien überhaupt sehr unempfindlich gegen elektrische Fel- der. So konnten sie z. B., an den Linien 4586, 4457, 4435, 4425, 4227 keine merkliche Zerlegung oder Verschiebung bis etwa 30000 Volt/cm feststellen. Auch bei anderen Linien war die Verschiebung nach rot sehr klein, am grössten bei den Dupletten der Hauptserie: 3737, 3706 und 3179, 3158 in einem Betrage von etwa 0,1s A.
1 Ann d. Phys. 43. 1914. S. 1031.
14 LICHTEMISSION A. D. OBERFLÄCHE V. KRISTALLEN
W. J. Humphreys,! der die Druckverschiebung der Calcium- linien untersucht hat, hat eine sehr kleine Verschiebung der meisten Linien nach rot bei einer Druckerhöhung gefunden. Bemerkenswert sind die Beobachtungen von Gale und Whitney,? die an den Linien 3644, 3630, 3624 und 3362, 3350, 3345 der ersten Calcium- tripletnebenserie eine Druckverschiebung nach kürzeren Wellen fanden.
Stark hat die Erklärung aufgestellt, dass die Druckverschiebung von zwischenmolekularen oder molelektrischen Feldern herrührt, dass also diese Verschiebung derselben Art ist, wie die durch äus- sere elektrische Felder hervorgebrachte. M. Ritter? hat weiter fest- gestellt, dass das Vorzeichen der Verschiebung durch die Wirkung des elektrischen Feldes ausnahmslos mit dem Vorzeichen der Ver- schiebung durch Druckerhöhung übereinstimmt, was für die Rich- tigkeit der von Stark gegebenen Erklärung spricht. So findet Ritter u. a., dass das Calciumtriplet 3362, 3350, 3345 nach kür- zeren Wellen um etwa 1,3 A verschoben wird bei einem Felde von
26000 Volt/cm, also in Übereinstimmung mit den Untersuchungen
von Gale und Whitney. |
Aus der Tabelle 1 geht hervor, dass alle Linien von 4586 Å bis 4227 Å scharf erscheinen. Dagegen zeigen die Linien des Triplets der ersten Nebenserie, 3644, 3630 und 3624 eine geringe Verbrei- terung nach violett. Das folgende Triplet hat zwar eine sehr ge- ringe Intensität, aber erscheint ganz unscharf und verwaschen. Diese Übereinstimmung im Vorzeichen mit den oben angeführten Beobach-
tungen deutet auf molelektrische Kräfte an der Kristalloberfläche.
Die Dupletlinien der Hauptserie sind alle nach rot verbreitert. Es ist zu vermuten, dass dieses von denselben molelektrischen Feldern herriihrt. Die Lichtstärke in diesen Dupletlinien ist gross, die Dispersion klein, also günstig.
1 Jahrb. d. Rad. u Elektronik 5. 1908. S. 324.
2 Astrophys. Journal 43. 1916. S. 165.
3
4 rn d. Phys.. 59. 1919. S. 170.
EEE I EEE
‘
ERGEBNISSE . 15
In der Tabelle 2 sind die Linien auf der Aufnahme: Wasserstoff- kanalstrahlen-Steinsalz zusammengestellt.
Tabelle 2. WASSERSTOFFKANALSTRAHLEN-STEINSALZ. Bezeichnung | Menenkupe Intensität Do Aussehen Na H.S.D. 5896 sehr gross 90 scharf Na 1N 5688 gross unscharf nach rot Na iN 4984 sehr klein stark verwaschen Hg 4860 klein 47 SEINES und verbreitert H, 4340 klein 99 verwaschen € 4270 klein scharf
Die D-Linie ist, wie aus der Tabelle hervorgeht, scharf. Das erste Duplet der ersten Nebenserie hat unscharfe Ränder, ist stärker nach rot als nach violett verbreitert. Das folgende Duplet ist zu schlecht definiert, als dass man darüber etwas aussagen könnte. In Über- einstimmung mit dem Unterschied zwischen den D-Linien und denen der ersten Nebenserie steht der Unterschied des Einflusses des elek- trischen Feldes auf die zwei Linienarten. Auf die D-Linien ist dieser Einfluss sehr kleiu,! dagegen ist bei der ersten Nebenserie nach Mes- sungen von Stark und Hardtke? eine Verbreiterung nach rot zu beobachten. | |
7. "ERGEBNISSE.
1:o Die von Kanalstrahlen getroffene Schicht von Flusspat und Steinsalz emittiert ein kontinuierliches Spektrum im Ultraviolett.
2:0 Linien von Kanalstrahlen und von Metallatomen, welche von Kanalstrahlen von einer Oberfläche abgetrennt werden, werden in einer unmittelbar an die feste Kristallflache grenzenden Schicht, dünner als 10 ° cm, so emittiert, als ob die Schicht teils dem festen
1 Vgl. R. Ladenburg, Die Einwirkung elektrischer Felder auf Absorptions- linien (D-Linien des Na-Dampfes). Physik. Zeitschr. 22. 1921. S. 549. 2 Ann. d. Phys. 58. 1919. S. 712.
Vv
16 LICHTEMISSION A. D. OBERFLÄCHE V. KRISTALLEN
Körper, teils dem angrenzenden Gasraum angehörte. Sie erschei- nen, wenn sie aus dieser Schicht heraustreten, verbreitert, wenn der Effekt des elektrischen Feldes auf sie gross ist, dagegen scharf, wenn dieser Effekt klein ist. Aus dieser Übereinstimmung der Verbreiterung mit dem elektrischen Effekte darf gefolgert werden, dass die Verbreiterung von molelektrischen Feldern des Kristalles bewirkt wird. |
Vorstehende Arbeit wurde im Wintersemester 1920—1921 im phy- sikalischen Institute der Universität Würzburg ausgeführt. Es war meine Absicht, die Untersuchung weiter auszudehnen und sie quan- titativ auszugestalten, ich wurde aber daran durch zwischengetretene Umstände verhindert. Immerhin diirften die bis jetzt gewonnenen Resultate trotz ihres qualitativen Charakters von Interesse sein.
Es sei mir auch hier gestattet, Herrn Prof. Stark für die Anre- |
sung zu dieser Arbeit und für die Bereitwilligkeit, mit der er alle Hilfsmittel des Institutes mir zur Verfügung stellte, meinen besten Dank auszusprechen.
Äbo, November 1921.
ACTA ACADEMIAE ABOENSIS MATHEMATICA ET PHYSICA I:6
UEBER AZETYLCELLULOSE AUS HOLZZELLSTOFFEN
VON
Prof. Dr. ERIK HAGGLUND, Mag. phil. NILS LOFMAN UND
Dr. EDUARD FARBER
ABO AKADEMI ABO 1921
a
Plt lees it N
Bee Wr
te 4
Hi
hi wu
INHALT.
TEL ND EJ TUT nn rs La a nern LSE AA LS EET ve TA 3.
NE
METHODEN ZUR DARSTELLUNG UND KENNZEICHNUNG VON AZE- TY ILI CHEL EIB) OOH BY OFS) Baa ay tena aT lab N a SV ES SA all A en SSG ER a ee The
2. FICHTENHOLZ ALS AUSGANGSMATERIAL. HERSTELLUNG DER] CELLULOSE AUS KONZENTRIERT SALZSAUREN SIE KENNEN SEE NN LES REA CLE AL RE LER RR ge A Ie VORBEHANDLUNG DER HOLZCELLULOSE MIT SCHWEFELSAURE- SAO RE RGA UREGEMISGE & 3... are. RUM amr nial denken 14. VORBERANDEUNGSMITVEISESSIG! öd ade De 14. SASSUERITZELEESTOFF ALS AUSGANGSMATERIAL . 2... 2... 22.02.20. 20. 15.
ASSBIETRIERPAPIER ALS “AUSGANGSMATERIAL u... 222.2... oc sel le 2222. 23.
EINLEITUNG.
Zu den chemischen Aufgaben, die im Kriege besondere Bedeutung gewannen, gehörte für Deutschland auch diejenige des Ersatzes von Baumwollcellulose. Er kam für. zwei Industriezweige in Betracht: die Herstellung von Schiessbaumwolle und diejenige der Azetylcellulose. Die Aufgabe schien vielleicht anfangs nicht besonders schwierig zu sein; eine ausgedehnte Fabrikation bestand ja für Zellstoffe aus Holz; sie konnte also das Ausgangsmaterial und viele Erfahrungen dazu bequem liefern. Aber dieses Ausgangsmaterial kann in chemischer und hier in Betracht kommender Beziehung merkwürdig verschiedenartiges Verhalten zeigen. Bei derartig hochzusammengesetzten Stoffen bezeichnet häufig noch das gleiche Wort chemisch verschiedene Stoffe; und wenn man zur weiteren Unterscheidung diejenige der Ausgangsmaterialien benutzt, so weist dies deutlich genug daraufhin, dass hier die chemische Charakteristik noch nicht vollständig gelungen ist. Man darf wohl selbst im Hinblick auf die neuste Veröffentlichung von Emil Heuser und E. Beedecker! sagen, dass die Frage nach der chemischen Eigenart von Cellulose verschiedener Herkunft noch nicht endgültig beantwortet ist. Andererseits braucht die technische Verwendung dieses Materials auf einen derartigen wissenschait- lichen Bescheid nicht zu warten; gerade sie liefert dann vielmehr Beiträge dafür, indem sie das Verhalten in der speziell verfolgten Richtung ge- nauer feststellt.
Sowohl bei der Herstellung von Schiessbaumwolle als auch bei der Azetylierung ergaben sich in der Tat besondere Verhältnisse, wenn statt der Baumwolle Cellulose anderer Herkunft verwendet wurde. Nach vielen, zum Teil Jahrzehnte zurück liegenden Untersuchungen sind die Schwierig-
1 Emil Heuser und E. Bedecker, Beiträge zur Kenntnis der Holzcellu- lose. Z. f. angew. Chem. 34. 461 (1921).
6 UEBER AZETYLCELLULOSE AUS HOLZZELLSTOFFEN
keiten für die erstere Fabrikation wohl überwunden worden!. Da erwies —
sich Sulfitzellstoff als geeigneter denn Natronzellstoff.
Uber die Azetylierung von Holzzellstoff hat jüngst W. Nauck? eine kurze Mitteilung veröffentlicht. Er betont darin den Einfluss der Ober- flächenbeschaffenheit und hält sie für massgebend für den Ausfall der Versuche; die Herstellung von Films gelang ihm.
Wir haben, vor einigen Jahren beginnend, derartige Azetylierungen ge- . nauer untersucht und können jetzt über die hauptsächlichsten der dabei er- haltenen Resultate berichten. Als Ausgangsmaterialien verwendeten wir Fichtenholz, Sulfitzellstoff, Filtrierpapier. Nach verschiedenartigen Vor- bereitungen dieser Stoffe wurden sie Azetylierungen nach bekannten und davon abweichenden Methoden unterzogen. Dabei stand das praktische Interesse im Vordergrunde; manchen neuen Erscheinungen konnte nicht in wissenschaftlicher Vollständigkeit nachgegangen werden. Es gelang schliesslich auch aus Sulfitzellstoff oder Filtrierpapier Präparate herzustellen, die als Lacke oder Filme brauchbar waren. Wir beschreiben im folgen- den aber auch diejenigen Versuche, die dieses Ergebnis nicht hatten, soweit dabei in anderer Hinsicht interessierende Beziehungen zu Tage traten. ;
1 Vgl. z. B. August Schrimpff, Nitrocellulose aus Baumwolle und Holzzell- stoffen. Sonderdruck aus der Zeitschr. f. d. ges. Schiess- und Sprengstoffwesen, 14. Jahrg. 1919 (J. F. Lehmanns Verlag, München 1919); darin ein historischer Rückblick, der den Anteil Wilhelm Wills wenig berücksichtigt. Ferner: S. D. Wells und V. P. Edwards, Die Verwendung von Holzzellstoff zur Gewinnung von Nitrocellulose. Paper, 12 Febr. 1919 (vgl. Cellulosechemie II 8 (1921).
2 W. Nauck, Beitrag zur Frage der Azetylierung von: Zellstoffen. „Cellulose- chemie‘, Wissenschaftl. Beibl. z. d. Zeitschr. „Der Papierfabrikant“, II 61 (1921).
Le I Zn Senn
1. METHODEN ZUR DARSTELLUNG UND KENN- ZEICHNUNG VON AZETYLCELLULOSE.
„Wohl auf keinem Gebiete der chemischen Technik dürfte der Umfang der geleisteten Arbeit, die Grösse der aufgewandten Kosten, die Mannigfaltigkeit der erfinderichen Tätigkeit, die Anzahl der Pa- tentanmeldungen und vor allem die Höhe der in die Entwicklung dieses Arbeitsgebietes gesetzten Erwartungen in solchem Missver- hältnis zu dem erzielten Resultat stehen, wie auf demjenigen der Azetylcellulosen.*! Es handelt sich ja bei der Azetylierung um einen Komplex von Reaktionen, die z. B. noch von mechanischen Einflüssen stark abhängig sind. Hat man das Azetylierungsgemisch selbst, aus Eisessig und Essigsäureanhydrid bestehend, festgelegt so bleibt noch die Wahl der Katalysatoren, der Temperatur und Zeitdauer für die eigentliche Veresterung, ausserdem auch die Wahl der Art wie man durch Hydrolysierung den Estern eine günstige Löslichkeit erteilt. Wenn so viele Faktoren variierbar sind, wenn die Zahl der „Freiheitsgrade“ anscheinend so ausserordentlich gross ist, gibt es natürlich eine Fülle von verschiedenen Versuchsmöglich- keiten. Temperatur und Dauer der Reaktion sind allerdings nicht, ganz unabhängig variierbar: wenigstens dann nicht, wenn man auf die Erzeugung von Triazetat, hinarbeitet. Bei zu starken Einwir- kungen geht die Veresterung weiter unter gleichzeitigem Abbau des Moleküls bis zu dem Cellobioseoktazetat. Aber lang vor seiner Bildung schon hört das Produkt auf, zu der vorgesehenen tech-
1 A. Eichengrün, Ullmann’s „Enzyklopädie d. techn. Chemie“ I. 114 (1914).
8 UEBER AZETYLCELLULOSE AUS HOLZZELLSTOFFEN
nischen Verarbeitung geeignet zu sein. „Die spröden amorphen Azetate, welche man bei zu lang dauernder Einwirkung des Azety- lierungs-Gemisches auf Cellulose in der Kälte erhält, die sich spie- lend leicht zu dünnen Flüssigkeiten in Chloroform lösen, verdanken ihre Entstehung anerkanntermassen der zu weit fortgeschrittenen Hydrolyse des Cellulosemoleküls“.! Ob man aber die „Azetylierung“
(wenn diese Bezeichnung für den ganzen Komplex der Reaktionen
gelten soll) an der geeigneten Stelle unterbrochen hat, das erkennt man erst bei der endgültigen Aufarbeitung. Besonders aufmerksam
muss man da den rascheren Verlauf bei den höheren Temperaturen
veriolgen. Da man es mit Reaktionen in heterogenen Systemen zu
tun hat, so kommt auf die mechanische Beschaffenheit der Cellulose.
sehr viel an.
Gross ist auch die Bedeutung der sogenannten Katalysatoren bei diesem Prozesse. Von all den als solche wirkenden Körpern ist ihre Fähigkeit, mit Wasser Verbindungen einzugehen, bekannt und darum wohl zunächst auch beim Versuche einer Erklärung ihres
Wirkens zuerst bedacht worden. Aber nach der Zusammensetzung
der Gemische kann es sich gar nicht um das Freiwerden von Was- ser handeln.” Die katalysierenden Zusätze sind zugleich auch für sich Lösungsmittel der Cellulose, oder sie stehen wenigstens solchen sehr nahe. So hat man denn auch diese Eigenschaft als ihre für den Azetylierungsprozess entscheidend wichtige hingestellt. Es dürfte sich aber damit wohl ähnlich verhalten, wie es für manche anderen katalytischen Prozesse E. Farber entwickelt hat. ? Der Katalysator, als ein Stoff der unter gewissen Bedingungen mit dem Ausgangs- material zu reagieren vermag, öffnet das Valenzield desselben in spezilischer Weise, und muss dann, u. a. auch seiner geringen Menge wegen, anderen anwesenden Koinponenten die eigentliche, zu der
1 H. Ost, Zeitschr. f. angew. Chem. 19. I. 997 (1906).
2 Vgl. dazu auch H. Ost, Zeitschr. f. angew. Chem. 32. 67 (1919).
3 E. Farber, Inaug. Diss. Leipzig 1916, S. 43 ff. (erscheint demnächst in Ann. d. Chem.); ferner auch „Die Naturwissenschaften“ 8. 324 (1920).
METHODEN ZUR DARSTELLUNG UND KENNZEICHNUNG 9
neuen Verbindung führende Reaktion überlassen. Unter diesen Ge- sichtspunkten gewinnt Ost’s Nachweis! einer anteiligen Bildung von Cellulose-Schweielsaure-Estern aus schwefelsäurehaltigen Ansät- zen besonderes Gewicht. Die so entstandenen Ester sind zumeist löslich in Chloroform, aber unlöslich, oder nicht „echt“, d. h. nicht wiederholbar löslich in Aceton oder Essigester. Dafür wen- det man nun die „Hydrolyse“ an: Man setzt nach Beendigung der Reaktion kleine Mengen Wasser, wässrige Säure oder Alkalilösung zu, und erhitzt einige Zeit lang. Auch diesen Vorgang hat Ost näher untersucht.” Er kennzeichnet ihn, übereinstimmend mit frü- heren Veröffentlichungen anderer Forscher, als eine teilweise Ver- seifung.
Die Durchführungen im einzelnen sind vor allem in zahlreichen Patentschriften beschrieben worden. Statt einer systematischen Auf- zeichnung > derselben sollen hier nur diejenigen erwähnt werden, die für unsere Untersuchungen Bedeutung gewannen.
Lederer geht nach D. R. P. 118538 (1899), 120713 (1899) und 163316 (1901) von Hydrocellulose aus, die für sich oder in dem Azetylierungsgemisch selbst erzeugt wird. Er erkannte auch den Ein- fluss der Temperatur auf die Eigenschaften des Produktes. Eichen- grün und Becker, die ebenfalls auf die Temperatur besondere Rücksicht nehmen, benutzen dagegen unveränderte Cellulose (D. R. P. 159524 [1901] Bayer), als Katalysator diente wie bei Lederer Schwefelsäure. Miles (Am. Patent 838350 [1905]) und die Firma Bayer (D. R. P. 252706 [1905]) haben dann die sogenannte hydra- tisierende Nachbehandlung erfunden, die zu einem in Aceton löslichen Produkte führt. Die Bad. Anilin und Sodafabrik verwendet nach D. R. P. 184145 (1904) feuchte Baumwolle. Knoevenagel
1 Vgl. Ost. Z. f. angew. Chem. 32. 69 (1919).
FILES Oo. th:
3 Vgl. dazu u. a. die auf S. 5 zitierte Abhandlung von Eichengriin, ferner etwa Edward C. Worden, Technology of Cellulose Esters, Bd. 8 New-York (1916), dazu auch ,Kunststoffe* //. 1—5 (1921).
10 UEBER AZETYLCELLULOSE AUS HOLZZELLSTOFFEN
(Knoll&C:o) hat sich u. a. in D.R. P. 203178 (1906) die Verwen- dung von neutralen, von Alkalisalzen verschiedenen Salzen, vor allem Chlorzink schützen lassen. (Vgl. ferner D. R. P. 303530 [1912]). Die Agia behandelt die Cellulose zuvor in einem Schwefelsäure-Sal- petersäuregemisch von grösserem Wassergehalte (D. R. P. 295889 [1911]). Alfred Wohl mischt und presst die Baumwolle mit Eisessig vor der Azetylierung (F. P. 448072 [1912]).
Zahlreiche weitere Abänderungen der Verfahren beim Azetylieren und darauffolgendem Hydratisieren sind in Patenten niedergelegt. An diesen Reaktionen die man im einzelnen noch so wenig durch-
schaut, ist eben immer das Endprodukt das Massgebend Interes-
sante gewesen. So gehört denn besonders zur Herstellung dieser Prä- parate diejenige „Uebung“, die immer, wenn sie in dieser Weise bedingend für den Erfolg ist, einen Mangel an wissenschaitlicher Durchdringung der Verfahren kennzeichnet.
Die analytische Charakteristik der Azetylcellulose bezieht sich natürlich auf den Azetyl-Gehalt in erster Linie. Dafür haben ja die Untersuchungen von Ost! die besten Methoden gegeben. Die alka- lische Verseifung liefert hier leicht Zersetzungsprodukte, die das Re- sultat störend beeinflussen; wir haben denn auch im folgenden die saure Hydrolyse nach Ost zur Azetylbestimmung angewandt.
Abweichend von andern Autoren geben wir ım folgenden stets „Azetyl“-zahlen an, d. h. den prozentualen Gehalt an CH, CO, während sonst oft auf Essigsäure, also CH, COOH berechnete Zahlen mitge- teilt werden.
Das Diazetat der Cellulose enthält 34,97 °/o CH, CO, entspr. “48. fo CH, COOH; für Triazetat sind die Werte: 44,8, bezw. 62,5 °/o; für Tetraazetat 52,1, bezw. 72,7.
Aber bei gleichem Azetylgehalte können Azetylcellulosen sehr verschiedenartig und verschiedenwertig sein. Es kommt nicht nur darauf an, ein Triazetat herzustellen; wesentlich ist ausserdem, in
1 Ost, Z. f. angew. Chem. 19. I. 996 (1906).
FICHTENHOLZ ALS AUSGANGSMATERIAL 11
welcher Art die Cellulose dabei verändert wurde. Das äussert sich in der Beziehung der Azetylcellulose zu verschiedenen Lösungsmit- teln und in der Viskosität der Lösungen. Während die geschicht- lich ersten Präparate hauptsächlich in Chloroform sich lösten, ging später das Bestreben — durch Miles zuerst erfüllt -- dahin, Aceton als Lösungsmittel zu benutzen. Eichengrün hat dann die Löslich- keit in Gemischen von Benzol und Alkohol entdeckt und gebraucht. Wir prüften auch das Verhalten gegen andere Flüssigkeiten und Gemische, wie sie beim Filmgiessen angewendet werden.
Ein weiteres analytisches Merkmal hat man in dem Reduktions- vermögen gegenüber Fehlingscher Lösung. Die „Kupferzahl“, näm- lich die von 100 g der veresterten Cellulose ausgeschiedenen Gramm Kupfer,! wäre natürlich bei einer ganz streng nur veresterten Cellu- lose kleiner als 1; in Wirklichkeit erhält man jedoch mehrfach grössere Zahlen, und zwar hat darauf nach unseren Versuchen die Endphase, die Hydrolyse der Azetylcellulose, keinen bemerkenswer- ten Einfluss. |
2. FICHTENHOLZ ALS AUSGANGSMATERIAL.
HERSTELLUNG DER CELLULOSE AUS KONZENTRIERT SALZSAUREN | LÖSUNGEN.
Aus den trockenen Sägespänen isolierten wir zunächst die Cellu- lose mit Hilfe von hochkonzentrierter Salzsäure. Nach 42 stündi- gem Stehen mit der fiinffachen Menge 40°/,-iger Salzsäure wurde das Filtrat in Wasser gegossen. Die ausgeschiedene Cellulose, von HCl und Wasser beireit, löste sich dann in verschiedenen Azetylier- ungsgemischen auf und gab stets ungeeignete Films, aber dies we- nigstens mit einigen bemerkenswerten Unterschieden.
Die Azetylgehalte waren auffallend niedrig, wenn nach den An- gaben von Miles Schwefelsäure als Katalysator gedient hatte: sie
1 Wir bestimmten sie nach dem Verfahren von Hägglund, Der Papierfabrikant, 17, 301 (1919).
12 UEBER AZETYLCELLULOSE AUS HOLZZELLSTOFFEN
stiegen nicht über 26 2/0! Chloroform löste die Azetylcellulosen nicht, wohl aber Aceton; diese Lösungen waren aber nur sehr wenig viskos. "5 g Cellulose, 12. g Essigsäureanhydrid, 20 g Eisessig wurden unter Kühlung mit 0, g conc. Schwefelsäure versetzt. Nach ver- schieden langer Aufbewahrung bei Zimmertemperatur erfolgte die Hydrolyse durch 2 g einer Mischung gleicher Teile von Wasser und Eisessig mit Zusatz von 5°/o Schwefelsäure bei 50° in 12 Stunden:
Dauer der Aufbewahrung nach Eintritt der Azetylgehalt Kupierzahl Auflösung 5 Stunden 26,0 °/o = 10 i ne — 20 25,8, 22,4 40 A 24,8 „ 14,3
Ein Vergleichsversuch mit reinster Watte führte zu einer guten Azetylcellulose. Ihr Azetylgehalt betrug 41,7 °/o, Cu-Zahl = 9,s; aus der viskosen Azetonlösung entstand ein klarer, zäher Film. -
Eine genaue Diskussion der Ursachen für den so spärlichen Ein- tritt der Veresterung ist vorläufig unmöglich; mit anderen Katalysa- toren gab dieselbe Cellulose wenigstens in Bezug auf die Menge des eingetretenen Azetyls normale Resultate; weiterhin wird noch zu beschreiben sein, dass eine etwas anders behandelte, durch Salzsäure aus Holz gewonnene Cellulose in einer anders zusam- mengesetzten Reaktions-Mischung ebenfalls höheren Acetylgehalt aufwies, obwohl dabei wieder Schwefelsäure den Katalysator bildete.
Mit Natriumäthylsulfat als Katalysator erhielten wir wieder nur unelastische Films, entsprechend der geringen Viskosität der Azeton- Lösungen. Aber die Azetylierung hatte doch offenbar einen ganz anderen Verlauf als mit Schwefelsäure genommen. Die Azetylzahlen blieben oft über 44°/o; sie sanken jedoch bei längerer Ausdehnung der Hydrolyse. Wir können damit auch an diesen Präparaten den
j 3 3 N
FICHTENHOLZ ALS AUSGANGSMATERIAL 13
Beiund von Ost an seinem ganz andersartigem Material bestätigen. Ansatz:
5 g Cellulose, 20 g Essigsäureanhydrid, 25 g Essigsäure, 0,5 g ‘ Natriumathylsulfat. |
Azetylierung I: bei 70° bis zur vollständigen Auflösung der Cellu- loser, X
Hydrolyse: Nach Zusatz von 4 ccm Wasser bei 55° und folgen- den Zeitdauern:
Hydrolysenzeit Azetylgehalt | Kupferzahl 5 Stunden 45,2 | 16,6
29 “i 44 3 16,0
ol X 43,5 | 16,6
Ansatz: wie soeben
Azetylierung Il: zuerst bei Zimmertemperatur; als nach 90 Stun- den jedoch keine vollständige Lösung erzielt war, wurde noch 25 Stunden lang auf 70° erhitzt.
. Hydrolyse: wie soeben
Hydrolysendauer Azetylgehalt | Kupferzahl 5 Stunden | 45,2 | 13,6 30 Z 43,9 19,4 100 5 | 39,4 14,4
Bei Ansätzen ohne Eisessigzugabe trat auch nach langer Einwir- kung keine Lösung ein. Aehnlich ging die Reaktion mit Chlorzink als Katalysator von statten. Mit zunehmender Hydrolysendauer nahmen die azetylierten Lösungen an Viskosität ab; die Azetylzahlen sanken von 47,7°/o auf 40°/o (nach 60 Stunden Hydrolyse bei 100°).
14 UEBER AZETYLCELLULOSE AUS HOLZZELLSTOFFEN -
VORBEHANDLUNG DER HOLZCELLULOSE MIT SCHWEFELSÄURE- ‚SALPETERSÄUREGEMISCH.
Nach den bisher mitgeteilten Versuchen hätte man schliessen können, dass die aus salzsaurer Lösung gewonnene Cellulose kein geeignetes Ausgangsmaterial für gute Films ist. Ehe wir diese Methode verliessen, wollten wir jedoch noch den Einfluss von Vor- behandlungen dieser Cellulose feststellen. Dafür liess sich z. B. das Verfahren der Agfa anwenden. Wenn danach die Cellulose durch verdünnte Nitriersäure „angeätzt“ und nach dem Auswaschen wieder getrocknet war, trat wohl mit Schwefelsäure als Katalysator keine sichtbare Azetylierung ein; aber schon mit verhältnismässig sehr wenig Chlorzink im Azetylierungsgemisch erhielten wir in normaler Zeit Auflösung der Cellulose. Allerdings entstand ein in den üblichen Medien unlösliches Produkt und angenähert dem Aze- tylgehalte des Tetraazetats: Diese Vorbehandlung hat also die Prä- paration ungünstig beeinflusst.
Ansatz:
5 g „vornitrierte“ Cellulose, 20 g Eisessig, 25 g
Essigsäureanhydrid, 0,5 g ZnCl,.
Azetylierung: 23 Stunden bei 50°.
Hydrolyse: mit 2 g Wasser 14 Stunden bei 50°.
Azetylgehalt: 49,0°/0; Kupferzahl 19,0.
VORBEHANDLUNG MIT EISESSIG.
Ein wenig besser verlief die Reaktion mit denjenigen Cellulosen, die zuvor dreimal mit Eisessig gewaschen waren. Allerdings ist damit nicht gemeint, dass nun Films von den gewünschten Eigen- schaften erzeugt werden konnten. „Doch löste sich wenigstens mit Schwefelsäure bezw. Natriumäthylsulfat als Katalysator die Cellulose im Azetylierungsgemisch auf; bei Zimmertemperatur waren dazu nur 2—3 Stunden erforderlich.
SULFITZELLSTOFF ALS AUSGANGSMATERIAL 15
Die Hydrolyse mit Wasser oder wässriger Kaliumhydroxyd-Lösung führte jedoch auch nach 50—60 Stunden bei 50° nicht zu Produkten von normaler Löslichkeit. Dem entsprach auch die zu geringe Viskosität der ursprünglichen essigsauren Lösungen.
Bei einer der Versuchsreihen hatte die vorbehandelte Cellulose die Kupferzahl 14,6. Die Ester daraus, die sich zumeist beim Fällen der ursprünglichen Lösungen in groben Flocken ausschieden, erga- ben z. B. 49,4°/o Azetyl, 17,7 Cu-Zahl; 48,0°/o Azetyl, 16,4 Cu-Zahl; 48,0°%/o Azetyl, 20,0 Cu-Zahl; 441°/o Azetyl, 14,s Cu-Zahl; 43,0°/o Azetyl, 14,1 Cu-Zahl.
Diese Ausfällungen lösten sich nur beim Erwärmen in Aceton oder Chloroform, am leichtesten noch in der Mischung aus 8 Tei- len Aceton und 15 Teilen Milchsäureäthylester („Acetonmischung‘“).
Andere Cellulosepräparate mit höheren und niedrigen Kupfer- zahlen verhielten sich im wesentlichen ähnlich. Man sieht daran nur wieder, wie sehr derartige analytische Bestimmungen der Er- eänzung durch andere bedürfen, wenn es sich um einigermassen vollständige Kennzeichnung handeln soll.
3. SULFITZELLSTOFF ALS AUSGANGSMATERIAL.
Sulfitzellstoff ist kein einheitlicher Stoff, Lignin und Pentosane, zu- weilen auch Hydro- und Oxycellulose, mineralische Bestandteile können darin in kleineren oder grösseren Mengen vorhanden sein.
Es ist nicht leicht, diese Fremdstoffe zu entfernen, ohne die Cellulose chemisch zu verändern. Die behutsame Einwirkung von Aetzalkali und von Oxydationsmitteln kann zum Ziel führen. Wir versuchten die Reinigung zunächst allein mit Oxydantien, wie Chlor- wasser, Permanganat, mit darauffolgender sorgfältiger Auswaschung der organischen und anorganischen löslichen Produkte. So wurde Sulfitzellstoff nach insgesamt 4-stiindiger Behandlung mit Chlor- wasser bei Zimmertemperatur, unterbrochen durch kalte alkalische
16 UEBER AZETYLCELLULOSE AUS HOLZZELLSTOFFEN
Auswaschungen, mit der Kupferzahl 5,69 erhalten; bei Verdoppelung der gesamten Einwirkungszeit sank dieser Wert nur wenig, auf 5,05, Die kurze Einwirkung von 5 °/o Königswasser führte zu einem Prä- parat von der Kupferzahl 2,10. Die 40-fache Menge 3 °/o-igen Wasser- stoffsuperoxydes, das in 100 ccm 10 ccm 24 °/o-iges Ammoniak ent- halt, blieb in anderen Versuchen einige Stunden in guter Beriihrung mit Zellstoff; die Kupferzahl betrug dann z. B. 3,42. 1 °/o-ige Kalium- permanganat-Lösung wirkte mit ähnlichen Ergebnissen ein. Auch vorheriges Behandeln mit Natronlauge löste nicht alle die Fehlingsche Lösung reduzierenden Substanzen aus dem Zellstoff heraus. Dabei wurden 100 g Zellstoff mit 400 ccm 2 ?/o-iger Natronlauge eine halbe Stunde lang geknetet, gründlich mit Wasser, dann verdünnter Salz- säure ausgewaschen, 40 Minuten lang Chlorgas durchgeleitet, während 3/ı Stunden mit 6 °/o-iger Natronlauge bei 45° behandelt und aus- gewaschen, gegebenenfalls auch mit Kaliumpermanganat und dann Natriumbisuliitlösung nachgebleicht. Ein Zellstoff, der vor dieser Reinigung 5,s als Kupferzahl hatte, wies danach den Wert von 4,63 auf.
Die Reduktionswerte blieben also im Vergleich zu Baumwolle noch ziemlich hoch. Die folgenden Versuche wurden zumeist mit durch Chlorwasser mit oder ohne vorherige alkalische Behandlung gerei- nigten Zellstoffen vorgenommen.
Bei den ersten zahlreichen Versuchen mit Sulfitzellstoff entstanden zumeist Produkte mit Azetylgehalten zwischen 48 und 52 ?/o, fast unabhängig von der Art des angewandten Verfahrens. Man muss annehmen, dass darin zum Teil Tetraazetate vorlagen. Die Versuche, mit Schwefelsäure zu azetylieren, brachten keine befriedigende Re- sultate, sowohl wenn nach Miles gearbeitet wurde, als auch bei einer an H. P. A. 7056 (1917) sich anlehnenden Abänderung, wo- bei das Essigsäureanhydrid, allein oder im Gemisch mit der kleinen Schwefelsäuremenge, erst nachträglich zugesetzt wurde. Auffallend hoch war der Reduktions-Wert gegenüber Fehlingscher Lösung: Neben Kupferzahlen von 13—14 lagen auch Werte wie 18,2,
SULFITZELLSTOFF ALS AUSGANGSMATERIAL 17
24 vor. Die Löslichkeiten waren gering und die Films brüchig. Diese Ergebnisse änderten sich zuerst nur in kleinen Einzelheiten und nicht in entscheidend günstiger Weise, als der Sulfitzellstoff zuvor mit Eisessig gewaschen und darauf getrocknet wurde. Wohl aber gelang es mit Chlorzink oder Natriumäthylsulfat (den Kataly- satoren, die Knoll nennt) gute Films herzustellen. Diese hatten auch sehr viel kleinere Kupferzahlen; die Azetylgehalte dagegen blieben manchmal noch in der Nähe derjenigen für Tetraazetate berechneten.
5 g Cellulose, mit Eisessig gewaschen und getrocknet;
20 g Eisessig, 25 g Essigsäureanhydrid,
les g Natriumathylsulfat:
Nach 22 Stunden bei 70° ist Lösung eingetreten; die nachfolgende Hydrolyse (mit Zusatz von 4 ccm Wasser) dauerte 29 Stunden bei 50°. Die Ausfällung, die in guter Ausbeute entstand, war hell und bildete Fladen.
Azetyl-Gehalt: 45,2 °/o; Kupfer-Zahl: 5,25. Chloroform löste nicht; Aceton, allein und im Gemisch mit Alkohol, gab beim Erwärmen Lösungen, aus denen klare, elastische Films entstanden.
Bei verlängerter Hydrolysendauer, nämlich nach 72 Stunden, waren die Löslichkeiten zwar annähernd dieselben geblieben; aber der Azetyigehalt war auf 43,4 %/o gesunken, die Kupierzahl auf 14,66 gestiegen. Die Films waren trotzdem noch ziemlich elastisch. © Das dabei verwendete Natriumäthylsulfat stammte von Merck (Darmstadt). Merkwürdig abweichend verhielt sich ein Präparat der Firma Kahlbaum (Berlin); die Auflösung der Cellulose im Azetylie- rungsgemisch dauerte nämlich in seiner Anwesenheit gelegentlich sechs Mal so lange wie mit jenem ersten Stoff und trat manchmal überhaupt nicht ein. Vermehrung des Essigsäureanhydrid-Anteils setzte die Reaktionsdauer beide Male herab. Aber der schlechtere Katalysator arbeitete immer noch vier Mal so langsam wie der an- dere. Beispielsweise wurden 5 g Cellulose mit 20 g Eisessig und den in der Tabelle angegebenen verschiedenen Essigsäureanhydrid-
18 UEBER AZETYLCELLULOSE AUS HOLZZELLSTOFFEN
Mengen bei 70” stehen gelassen; die Menge des Katalysators blieb stets 1,2 g in jedem Ansatz. Nach der Auflösung wurde mit Zusatz von 4 ccm Wasser 48 Stunden bei 50° hydrolysiert.
Essigsäure- Präparat Merck: Präparat Kahlbaum: anhydrid | Auflöse-Zeit Löslichkeit | Auflöse-Zeit Löslichkeit 3 Stunden in Aceton Stunden in Aceton 25 64 gut 410 trüb 1ösl. 200 35 gut 190 unvollstandig löslich 50 17 gut 68 wenig lösl.
Auch mit dem ersten Präparate ist hier die Auflösungszeit beträcht- lich. Sie konnte durch gute Verteilung der Cellulose und häufiges Bewegen des Ansatzes, besonders auch durch Einsaugen der Azety- lierungsmischung in die evakuierte Cellulose, stark herabgesetzt wer- den, z. B. auf 24 Stunden. Der Unterschied in der Wirkungsweise der beiden Katalysatoren blieb dennoch bestehen, ja er trat noch verstärkt zu Tage. Daran änderte sich auch nichts, als statt prak- tisch ganz trockener Cellulose solche von 3 und 6 °/o Wasser-Gehalt verwendet wurde.
Von Natriumbisulfat reichten 0, g, zu der Mischung von 5 g Cellulose, 20 g Eisessig und 25 g Essigsäureanhydrid zugesetzt, nicht aus um eine Auflösung zu erzielen. Wenn ausserdem noch 0,5 g Natriumäthylsulfat Kahlbaum anwesend war, erhielt man nach 206 Stunden (70°) Azetylierung und darauffolgender Hydrolyse einen in Aceton und Acetonmischung schwerlöslichen Ester und brüchige Films. Besser wirkte die Mischung mit 0,2 g Natrium- bisulfat + 0,5 g Natriumäthylsulfat, doch war die Auflösedauer noch immer ähnlich hoch. Vom Natriumbisulfat allein genügten auch 0,5 g nicht; erst mit 1 g davon war nach 28 Stunden (70°) Lösung eingetreten, und die hydrolysierte Azetylcellulose gab aus Aceton- mischung einen guten Film.
SULFITZELLSTOFF ALS AUSGANGSMATERIAL 19
Die Azetylierung ist auch insofern abhängig von der „Vorgeschichte“ der Cellulose, als gewisse Vorbehandlungen mit einzelnen Kompo- nenten der Azetylier-Gemische oder anderen organischen Flüssig- keiten die Dauer der Veresterung, gelegentlich sehr stark, beein- flussen. Wenn die trockene Cellulose mit Eisessig vermischt meh- rere Stunden lang (z. B. 24) bei gewöhnlicher Temperatur aufbe- wahrt wurde, so machte sich kaum ein Einfluss auf die nachfolgende Reaktion bemerkbar. Geschah die Aufbewahrung bei 70° ca. 18 - Stunden lang, so ging die Azetylierung mit Natriumbisulfat (0, g +0, g Wasser auf den üblichen Ansatz mit 5 g Cellulose) als Katalysator unverkennbar beschleunigt vor sich; die Auflösung im Azetylierungsgemisch erfolgte in durchschnittlich 4 Stunden bei 70°, während sie sonst eine etwa 3—6 Mal so lange Zeit beanspruchte.
Sehr hindernd wirkte aber die Vorbehandlung mit dem Gemisch von Eisessig und Essigsäureanhydrid bei hoher Temperatur. 20 g Cellulose wurden mit 80 g Eisessig und 120 g Essigsäureanhydrid am Rückflusskühler gekocht; dabei war natürlich für Fernhaltung von Feuchtigkeit gesorgt. Nach dem Abkühlen kamen 2, g Nat- riumathylsulfat (von Merck, Darmstadt) als Katalysator hinzu. Hatte das Kochen 1!/2 Stunden gedauert, so vergingen 116 Stunden bis zur Auflösung, die sonst bei derselben Temperatur (70°) etwa 12 Stunden dauern kann. 2-stündiges Kochen verzögerte die Aze- tylierungsdauer noch mehr, und nach 3 Stunden langem Sieden trat überhaupt keine Auflösung mehr ein.
Zur Erklärung könnte man zunächst daran denken, dass die Er- hitzung in diesem Gemisch zu einem Abbau der Cellulose geführt hätte. Ein solcher ist nicht nachweisbar, wenigstens wenn man die Kupferzahlen als Kriterium benutzt. Merkwürdigerweise wird die starke Reaktionsverzögerung zum Teil aufgehoben, wenn man nach dem Kochen ganz wenig Wasser zusetzt. Proben von je 5 g Cellulose wurden mit 20 g Eisessig und 30 g Essigsäureanhydrid 2 Stunden gekocht; nach dem Abkühlen kamen je 1,5 g Natrium- äthylsulfat hinzu, und in die eine der Mischungen ausserdem noch
+
20 UEBER AZETYLCELLULOSE AUS HOLZZELLSTOFFEN
0,1 ccm Wasser. Nach dreitägiger Aufbewahrung war nur in der mit Wasser versetzten Probe Auflösung eingetreten. !
Uebrigens wirkte auch vorheriges Kochen mit Chloroform, Ace- tylentetrachlorid oder Alkohol ähnlich wie das Kochen mit Eisessig- Essigsäureanhydrid-Gemisch. Eine Cellulose mit der Kupferzahl 5,02 behält diesen Wert auch bei, nachdem sie eine halbe Stunde lang mit einer der genannten organischen Flüssigkeiten gekocht worden war.
Die nachfolgende Azetylierung mit Natriumbisulfat als Katalysator, die
sonst bei 70° in etwa 10 Stunden zu der Auflösung der Cellulose führte, beanspruchte hier 140—200 Stunden bei derselben Tempe- ratur.
Anders wirkten diejenigen Vorbehandlungen, bei denen die Cel- lulose zunächst mit den entsprechenden Mengen von Eisessig und Katalysator zusammengebracht wurde. Man liess dann die Mischung
mehrere Stunden je nachdem bei Zimmertemperatur oder bei 50—70° ©
stehen und fügte danach erst die nötigen Mengen Essigsäurean- hydrid zu.
Wenn dabei Schwefelsäure der Katalysator war, so bedingte die Dauer der Vorbehandlung diejenige der Auflösung im Azetylierungs- gemisch. 5 g Zellstoff wurde mit 20 g Eisessig und 0,2 g konc. Schwefelsäure bei Zimmertemperatur bezw. 16, 20, 50 Stunden auf- bewahrt. Nach Zusatz von 25 g Essigsäureanhydrid dauerte dann die Auflösung bei Zimmertemperatur entsprechend 30, 25, 20 Stunden. Die hydrolysierten Produkte hatten Azetylgehalte von etwa 46 ?/0; die Kupierzahl derselben war von etwa 4 auf etwa 7 gestiegen; die ‚Löslichkeit in den üblichen Medien war aber nicht immer günstig, und der Film liess zu wünschen übrig.
1 Letzthin ist ein Patent des Vereins für Chemische Industrie in Mainz, Frank- furt a/Main D. R. P. 339824, Kl. 12 0, Verfahren zur Herstellung von Cellulose- azetat, erteilt worden, nach welchem man zur Erzielung eines wasserfreien Ausgangsmaterials die Mischung von Cellulose, Essigsäureanhydrid, Eisessig oder einem anderen Verdünnungsmittel längere Zeit einwirken lässt, ehe man den zur Veresterung erforderlichen Katalysator hinzufügt.
RE a Ze
SULFITZELLSTOFF ALS AUSGANGSMATERIAL 21
Versuche mit einer vorher „hydrolysierten“ Cellulose nach dem Verfahren von Lederer führten nicht zu guten haltbaren Films; Ost beschreibt! das Verhalten von Watte bei zweistündiger Einwirkung von Eisessig (achtiache Menge) und Schwefelsäure (40 °/o vom Cellulosegewicht); die Watte war dann stark briichig geworden, ohne dass die Elementaranalyse eine Veränderung sicher zu erken- nen gestattete. Wir liessen Eisessig und Schwefelsäure in demselben Verhältnis wie angegeben verschiedene Zeiten lang auf Sulfitzellstoff einwirken. Die Flüssigkeit wurde dann abgesaugt und die Cellulose durch Zugabe der üblichen Mengen Eisessig und Essigsäureanhydrid bei Zimmertemperatur azetyliert. Nach der Auflösung wurde dann in der einen Serie sofort mit Wasser gefällt, in der anderen vorher erst 12 Stunden bei 50° hydrolysiert.
Enorsaleise hiner
ey Azetyl | Kupfer-Zahl Aceton- cellulose Gehalt Aceton Chloroform | oder Ester- | mischung? 1) Dauer der Vorbehandlung: 17 Stunden » » Auflösung: 31/2 Stunden icht hydro- . en a 46,26 | 5,91 nicht wenig | nicht hydrolysiert| 43,75 | 27 Al | gut | gut | gut 2) Dauer der Vorbehandlung: 24 Stunden 3 „ Auflösung: 14/2 Stunden icht hydro- tr in d nen ats Ried. 37,11 | 9,91 Ne kl unlöslich | unlöslich hydrolysiert 64,55 | 20,69 | gut | gut | gut
1 Zeitschr. f. angew. Chem. 19. 994 (1906). 2 Die „Estermischung“ besteht aus 85 Teilen Ameisensäure-Äthylester und 15 Teilen Milchsäure-Äthylester.
22 UEBER AZETYLCELLULOSE AUS HOLZZELLSTOFFEN
EGS ba cab kee at
Azetyl- Azetyl- Ku pfer-Zahl Aceton- cellulose Gehalt Aceton Chloroform | oder Ester- mischung
3) Dauer der Vorbehandlung: 48 Stunden As „ Auflösung: 20 Minuten
capes 29,09 | 10,41 | unlöslich | unlöslich | Er hydrolysiert 57,02 | 30,53 | gut | gut | — 4) Dauer der Vorbehandlung: 72 Stunden 5 „ Auflösung: 10 Minuten ee 41,36 | 4,06 unlöslich | unlöslich — hydrolysiert 51,64 | SONG | gut | gut | —
Je länger die Vorbehandlung dauerte, umso rascher vollzieht sich
die Auflösung. Aber die eigentliche Veresterung ist dann nicht
immer auch schon beendet; gelegentlich hat sie noch nicht einmal
die Stufe des Diacetats erreicht. Die darauffolgende „Hydrolyse“ ist dann zugleich auch noch weitere Veresterung; und gerade dann schiesst sie zugleich auch noch über das Triacetat hinaus, trotzdem das 12-stündige Erhitzen auf 50° eine milde Hydrolysierbedingung darstellt. Immer wächst der Reduktionswert gegen Fehlingsche Lösung ausserordentlich stark an. In allen Fällen stellten die Ver- dunstungsrückstände unbrauchbare, zum Teil schon beim Trocknen zerspringende Haute dar.
Von allen genannten Vorbehandlungsweisen hatte sich also dieje- nige mit Eisessig allein noch am besten bewährt. Noch günstiger und zuverlässiger wirkt. es, wenn dem Eisessig gleich auch als Azetylierungskatalysator das Natriumbisulfat oder Chlorzink zuge- fügt wird. Man arbeitet dann in folgender Weise:
FILTRIERPAPIER ALS AUSGANGSMATERIAL 23
Zu 5 g Cellulose wird eine Lösung von 20 g Essigsäure (100 °/o), 0,5 g Wasser und 0, g Natriumbisulfat gesetzt. Diese Mischung wird während 17 Stunden bei 50—70° erwärmt. Danach rührt man in die gekühlte Mischung 25 g Essigsäureanhydrid ein. Die Tem- peratur darf dabei nicht über 60° steigen. Man erhält auf diese Weise eine viskose Lösung. Durch Erwärmung der Reaktions- mischung bei 70° während kurzer Zeit (ungefähr eine halbe Stunde) wird die Reaktion beendet. Dann wird etwas abgekühlt. Man setzt 5,5 —6 ccm Wasser zu und lässt die Mischung während 65—70 Stunden bei 50° stehen. Durch Eingiessen der Lösung in überschüssiges Wasser wird ein in Chloroform unlösliches, aber in Aceton lösliches Produkt ausgefällt.
4. FILTRIERPAPIER ALS AUSGANGSMATERIAL.
Über Azetylierungen von Filtrierpapier hat Ost 1906 einige An- gaben veröffentlicht. Er scheint bei seinen Versuchen keine Beson- derheiten gegenüber der Watte als Ausgangsmaterial gefunden zu haben.
Wir wandten auf das Filtrierpapier (Schleicher-Schüll) dieselben Verfahren an, die wir mit dem gereinigten Sulfitzellstoff durchpro- bierten. Zunächst also versuchten wir die aus der salzsauren Lösung durch Wasser ausgefällte Cellulose zu verwenden. Die Kupferzahlen der Papiere schwankten in verschiedenen Proben zwischen 1 und 3. Wenn die in der zehnfachen Menge 40 °/o-iger Salzsäure aufgenom- mene Cellulose nach 10 Minuten langem Verweilen in der filtrier- ten Lösung mit Wasser gefällt wurde, kam sie mit einer Kupfer- Zahl von 17,7 unkorr., 13,9 korrigiert wieder heraus; bei einstün- diger Einwirkungsdauer der Salzsäure stieg dieser Wert auf 28,0, bezw. 22,s, als wahre Kupferzahl. In entsprechender Höhe hielten sich die Reduktionswerte der damit erzeugten Azetylcellulosen, und wie immer in solchen Fällen ergaben sich nur spröde Films. Dabei können die Azetylgehalte bei verschiedenen Ansätzen zwischen den
24 UEBER AZETYLCELLULOSE AUS HOLZZELLSTOFFEN
Werten für Di- und für Tetraacetat schwanken. Zum Beispiel wurden 5 g der wie oben beschrieben erhaltenen Cellulose von der Kupfer- zahl 22,3 nach Auswaschen mit kaltem Eisessig getrocknet und mit 20 g Eisessig, 25 g Essigsäureanhydrid und 0,2 g conc. Schwefel- säure gemischt. Nach 6 Stunden bei Zimmertemperatur war Auf- lösung eingetreten. Darauf folgte die Hydrolyse mit 4 g Wasser bei 50°. Dauerte sie 48 Stunden, so zeigte die Ausfällung einen Azetylgehalt von 50,2 °/o und die Kupferzahl 20,2; nach 60 stündiger Hydrolyse einer anderen, im übrigen genau so behandelten Probe war der Azetylgehalt 37,3 °/o, die Kupferzahl 19,6s. | Auch für Filtrierpapier erwies sich schliesslich diejenige Arbeits- weise als sicherste, welche wir oben für Suliitzellstoff angegeben haben: Vorbehandlung mit Eisessig und Natriumbisulfat, bezw. Chlorzink, möglichst kurze Azetylierung bei 65°, Hydrolyse während dreier Tage bei 65—70”. Sie führt dann zu Triacetaten mit niedri- gen Kupferzahlen, aus deren Lösungen in der Acetonmischung gute, elastische, dauerhafte Films gewonnen werden können.
Wissenschaftliches Laboratorium der Th. Goldschmidt A. G., Essen und Mannheim—Rheinau.
ACTA ACADEMIAE ABOENSIS MATHEMATICA ET PHYSICA I: 7
UBER EINE VERALLGEMEINERUNG DES EUKLIDISCHEN ALGORITHMUS
VON
NILS PIPPIN G
DR. PHIL., LEKTOR DER MATHEMATIK AN DER AKADEMIE ABO (FINLAND)
ABO AKADEMI ABO 1922
Dips a
N
BO TRYCKERI OCH
UBER EINE VERALLGEMEINERUNG DES EUKLIDI- SCHEN ALGORITHMUS.
1. Wir betrachten ein System von drei positiven reellen Werten en Sn
Falls sie alle unter einander verschieden sind, bilden wir aus dem gegebenen System zwei neue Systeme nach der folgenden Regel.
Das u:te System (u= 1,2) geht hervor, indem .der grösste Wert, ty, durch den Wert t,—t, ersetzt wird, während die Werte t,, t, unverändert gelassen werden.
Die so gebildeten Systeme bestehen aus drei positiven Werten; falls jedes derselben nur unter einander verschiedene Werte enthält, wiederholen wir dasselbe Verfahren in Bezug auf alle beide. Hierbei erhalten wir 2” neue Systeme, u.s.w. Bei dem v:ten Schritte gehen 2” neue Systeme hervor.
So lange kein einziges von den betrachteten Systemen zwei gleiche Werte enthält, setzen wir das genannte Rekursionsverfahren fort; die bei dem folgenden Schritte hervorgehenden Systeme bestehen dann aus lauter positiven Werten. Falls wir aber ein System bekommen, das zwei gleiche Werte enthält, brechen wir das Verfahren ab; in diesem Falle sagen wir daher, dass der Algorithmus abbricht, sonst lässt er sich ins Unendliche fortsetzen.
Es sei w eine positive reelle Grösse, und wir wenden den obigen
4 ÜBER DEN EUKLIDISCHEN ALGORITHMUS
Algorithmus auf die Potenzen 1, ,? an. Hierbei erhalten wir eine Kette von Systemen
ea er ma! obi eile el a,e aise) « ee es eivehalelera ieilee en je! «
welche wir die zu w gehörige allgemeine Kette 2:ter Ordnung nen- nen. 4
Im folgenden setzen wir diese Kette als unendlich voraus, und wir bezeichnen sie kurz mit X. Von der Kette X greifen wir eine Reihe von Systemen
in der Weise heraus, dass jedes System si) eines der beiden Sys- teme ist, welche aus dem System Se hergeleitet wurden.
Diese Kette nennen wir eine Unterketie der allgemeinen Kette K. Wir wollen darlegen, dass die von Jacobi angegebene Verallgemei- nerung des Euklidischen Algorithmus eine Reihe von Systemen gibt, die sämtlich in derselben Unterkette der Kette K vorkommen.
Mit Hilfe der Gleichungen
(a) rei Yen) Coa, Com a ed ee (v = 0,1, 2,---),
in denen /, und m, die grössten ganzen Zahlen bedeuten, welche =U, U, bzw. w,:u, sind, leitet Jacobi — wie bekannt — aus den Ausgangswerten u,=1, W=o, W,=w? sukzessiv neue Wertesys- teme her.
1 Eine allgemeine Kette m:ter Ordnung ergibt sich, wenn wir von dem Werte- system 1, w, w”, .. , m” ausgehen, aus demselben nm neue Systeme bilden, aus diesen wieder n? neue Systeme, u. s. w. (Vgl. Nils Pipping: Ein Kriterium für die reellen
algebraischen Zahlen, auf eine direkte Verallgemeinerung des Euklidischen Algo- rithmus gegründet; Acta Academiae Aboensis, Mathematica et Physica I, Äbo, 1921.)
ÜBER DEN EUKLIDISCHEN ALGORITHMUS 5
Es lässt sich ohne Mühe zeigen, dass falls das System u,, v,, w, in der Kette K vorkommt, so ist dasselbe mit dem System (a) der Fall. ;
Hierbei sind drei Möglichkeiten zu unterscheiden.
1:o. Von den Werten UU, Op W, ist u, der grösste.. Dieser Fall kann nur für »=0 zutreffen. Wir haben dann u, >v,, u, >w,, so dass /,=m,==0; das System (a) ist mit dem System u,, v,, w, identisch und kommt mithin in der Kette K vor, w. z. b. w.
2:0. Von den Werten U, V,, w,ist u, der mittlere. Es ist dann W, >, > %,; denn falls wir v, >u, >w, hätten, würde aus der letzteren von diesen Ungleichungen folgen, dass »=0 ware, d. h. u, wäre entweder der grösste oder der kleinste von den Werten U, V,, w,. Die Werte des Systems (a) sind daher
N EN NS ile,
weil ja !, =0 aus der Ungleichung u, >, hervorgeht. Dieses Werte- system kommt in der Kette X vor. Es ist nämlich w, der grösste von den Werten w,,v,,w,; das System u,,v,, w,—u, kommt mithin in der Kette X vor, womit unsere Behauptung im Falle m,=1 be- wiesen ist. Für m, >1 ist w,— u, der grösste von den Werten U, V,, W,—U,, WOTaus folgt, dass die Werte u, U, w,—2 u, ein System der Kette K bilden. Vermittels vollständiger Induktion lässt sich auf dieselbe Weise schliessen, dass allgemein das System U,» dv, W,—m, u, in der Kette X vorkommt, w. z. b. w.
3:0. Von den Werten u,,v,, w, ist u, der kleinste. Wir subtra- hieren von dem grössten der Werte u,, v,, w, den Wert z,, von dem grössten Wert des so erhaltenen Systems subtrahieren wir wie- der den Wert u,, u. s. w. Dieses Verfahren gibt uns nach (l,+m —2) Operationen das System
v,—l,—1)u,, w,—(m,—1)u,, U,»
in welchem die beiden erstgeschriebenen Werte noch immer > u,
sind. Der Wert z, von dem grösseren der beiden Werte v,—(/,—l)u
v?
6 ÜBER DEN EUKLIDISCHEN ALGORITHMUS
w,—(m —1)u, subtrahiert, gibt ein System mit dem zweiten derselben
als grösstem Wert. Wenn wir von diesem noch den Wert u, subtra-
hieren, erhalten wir endlich das System (a), welches also in der Kette K vorkommt, w. z. b. w. Aus den obigen Erwägungen geht ferner folgendes hervor: das
System (a) stammt sozusagen von dem System u,, v,, w, her, d.h.
das System (a) ist eines von den Systemen, welche entweder direkt oder durch gewisse Zu en ale Systeme aus dem System U, v, w, hergeleitet wurden. Weil 5, a) a, ao") — (U, Ope a) = en wir mithin das Ergebnis gewonnen:
Die Wertesysteme der Jacobi-Kette kommen sämtlich in derselker Unterkette der Kette K vor.
2. Die Kette K wird periodisch genannt, falls es in derselben eine Unterkette gibt, welche zwei Systeme enthält, deren Werte pro- portional sind.
Nach dem in Nr. 1 dargelegten fällt mithin die zu » gehörige allgemeine Kette X dann immer periodisch aus, wenn die zur frag- lichen Grösse w gehörige Jacobi-Kette einen periodischen Verlauf hat. Bezüglich aller Beispiele, für welche eine periodische Jacobi- Kette hergeleitet wurde, lässt sich daher sofort schliessen, dass auch die Kette X periodisch ist.
Ferner wollen wir zeigen, dass falls die zu » gehörige allgemeine Kette (2:ter Ordnung) periodisch ausfällt, so ist w eine algebraische Zahl 3:ten Grades.
Diejenige Unterkette, welche unserer Annahme gemäss zwei Sys- teme enthält, deren Werte proportional sind, bezeichnen wir mit
Sy Se Se jai
uud es seien u, dh @, die drei Werte, welche dem System Ay! ge- hören. Wir dürfen hierbei
Uj=1, %,=o W=o
a ae EN SE nee
ÜBER DEN EUKLIDISCHEN ALGORITHMUS 7
wählen, und allgemein: wenn wir aus dem System 5, , das Sys- tem S, herleiten, ersetzen wir den grössten Wert jenes Systems durch einen neuen Wert nach der in Nr. 1 angegebenen Regel ohne die gegenseitige Reihenfolge der fraglichen Werte zu verändern; u,, V,, w, bezeichnen dann eindeutig definierte Werte. Es lassen sich die Werte des Ausgangssystems S, in der Form w= k=l O.a+0.07 =p, +p, @ + py w? VY = 0 =0+1-9+0-0?=9,4+ 4) 9+4q,'
MW =o?=04+0-04+1-e?=7,4+ 7,04 7," wo
schreiben, und falls wir sukzessiv mit sämtlichen Koeffizienten p, g, r dieselben Operationen ausfiihren wie mit den entsprechenden Werten u, v, w, erhalten wir daher die Gleichungen
Dp, =D, 0 Dh 0)” Gig Dy Ora ae vy
0 + Ben w?
|
U, (1) v, 2,
mit ganzzahligen Koeffizienten p, q, r. Dem obigen gemäss ist die Determinante dieses Gleichungssystems gleich +1, und wir erhalten mithin
in, u, a Q, RO), (2) on Pa Be OO TR, Ww,
DER " " t " a in, u, ci Q, U, A R, Wy»
wenn wir wie gewöhnlich die Unterdeterminanten mit P,, Q,, R,, Seer) ies, ©, Oo, ie,” ‚bezeichnen. Es seien die Werte u,, v,, w, der Grösse nach geordnet « ‚>ß,>y,, wobei 2, a,+ a, w+ a, w? ' u 2 (1) P,= b, ac b, 0 + b, 0)
wet r IE 2 fn ka org are, a.
8 ÜBER DEN EUKLIDISCHEN ALGORITHMUS
Wenn wir dieses Gleichungssystem auflösen und die Unterdeter- minanten mi A, BA, CA, Dr AB ec bezeichnen, erhalten wir [vgl. (2)]
ae = Ay Oy Dae Ci (2) EO A, yt Bg Cty +w? =A,'«, SE BY ß, üb © 7,9 in welchen Gleichungen die linken Glieder entweder alle drei positiv oder alle drei negativ sind. Nehmen wir jetzt an, es sei «,:8,:7, = ar: B,:7,; (1 >), So dass
(3) ee 2 = 1-9; 0<0<1).
a, Py Vr
Nach (2)’ und (3) erhalten wir dann +9= A, en Dell: C,1,=a+bo+co? + 90 = A, Cat Bein ya oo. 0 +90= A, oe +B,P+C/y,=g+ho-+ios,
wo a, b, c, d, e, f, g, h, i ganze Zahlen bedeuten, und nach Eli- mination von © gehen daher zwei kubische Gleichungen mit ganz- zahligen Koeffizienten hervor:
cw + (b—f)w? + (a—e)w—d=0 fo? + (e—i) w + (d—h)wo—g=0.
Beide können nicht Identitäten sein. Denn in diesem Falle wäre c=d=f=-g=b=h=0; a=e=i=+@, d. h. © wäre eine ganze Zahl, was mit (3) in Widerspruch steht.
Wir haben mithin bewiesen, dass w eine algebraische Zahl, höch-
stens vom 3:ten Grade ist. Weil aber die betrachtete Kette — un-'
serer Annahme gemäss — nicht abbricht, kann » keine algebraische Zahl 1:sten oder 2:ten Grades sein (vgl. die auf Seite 4 zitierte Abhandlung), woraus folgt, dass » tatsächlich eine kubische Irratio- nalzahl ist.
ÜBER DEN EUKLIDISCHEN ALGORITHMUS 9
3. O. Perron’ hat wie bekannt gezeigt, dass die Jacobi-Kette _ konvergiert. Wir fragen jetzt, ob dasselbe mit einer beliebigen Un- terkette der Kette X der Fall ist. Wir fragen m. a. W., ob allgemein die Beziehungen
P , / R , ee Ti rei 0.40 Im, lg (4) SR, aes alas: " | Q " R "HÉ a ee Um 2002 2
. ‚gültig sind.
Es ist leicht darzutun, dass dem so nicht ist. Zu dem Ende brauchen wir nur den Zusammenhang zwischen den Zahlen |e ice R, und den Zahlen Pu ER Rau eh zu untersuchen.
Nehmen wir einen Augenblick an, dass die Werte “UO der Grösse nach geordnet w, >v, >u, sind, und dass wir als Subtra- henden den Wert v, wahlen, um aus dem System S, das System S,41 herzuleiten. Wir haben dann
" "|
RT P,+ı Py Piren ge ESS 2» A a Gen red dry Dye ey
' ER / IDS yf " = cd meme Cp) Oy ia ye a Sey Dey ay ry ET
und daher LA Sia as Fee i tae PA lee Oy: Lor Q, ne R, Qt ine Q', i IN AT Il Q, a I Ie ee ea Ae Bo AGNE Ra eR’, -
Ähnliche Beziehungen erhält man, falls die Werte u,,.0,,@, der Grösse nach geordnet in einer anderen Reihe aufeinanderfolgen, oder falls wir als Subtrahenden nicht den mittleren sondern den kleinsten von den Werten ui, U,, W, wählen. Wir sehen allgemein, dass
1 Siehe Mathematische Annalen, Bd. 64, 1907.
10 UBER DEN EUKLIDISCHEN ALGORITHMUS å
von den Zahlen P,,---, R,” nur diejenige, welche in den Glei- chungen (2) Koeffizienten des Subtrahenden z,, v, oder w, sind, verändert werden.
Bei der Wahl des Subtrahenden sind stets zwei Möglichkeiten offen. Wir können folglich für jeden Index » z. B. den Wert u, als Subtrahenden vermeiden. Es ist dann
py?
P,=P,=1, P' =P,' =0, P,” =P,’ =0; (v =0,1,2,---),
und wir haben folglich nicht
/ " lm or lim es ar
P P
v=o Vv v— 9° Vv 4. Betrachten wir aber speziell diejenige Unterkette
So Sig ng een
welche sich ergibt, wenn wir bei jedem Schritt den mittleren der Werte u,, v,,W, von dem grössten derselben subtrahieren; fiir die so gebildete Kette sind die Beziehungen (4) tatsächlich gültig, was wir nunmehr beweisen werden. Zu dem Ende schreiben wir die Werte u,» v,, stehende Tabelle I auf, und daneben schreiben wir eine andere Tabelle II auf, worin «,', al”, al (»=0,1,2,---) eine gewisse Per- mutation der Ziffern 1, 2, 3 angibt; dem grössten (bzw. dem mitt- leren und dem kleinsten) von den Werten w,, v,, w, der Tabelle I entspricht auf demselben Platz in der Tabelle II die Ziffer 1 (bzw.
2 und 3).
w, in die unten- -
Tabelle I. Tabelle II. 1 2 3 uy, Uy Wo oF oye at (1) (2) (3) uU, U, Wi 0 Oy Oy Uo Uy We PE AN
ÜBER DEN EUKLIDISCHEN ALGORITHMUS 11
_ Nach dem Bildungsgesetz der betrachteten Kette bleibt — mit Rücksicht auf das Axiom von Archimedes — kein Wert u, v oder ww fortwährend der kleinste, sondern es gibt sicher einen Index » derart, dass wenigstens einer von den Werten uw, su, als der kleinste von den Werten u,
1 7 Dy kleiner ye ON Oe ASE. nd weil bei der a jedes neuen Systems nur einer von den Werten z, N verändert wird, hat nur einer von den Werten 00, Oy aie genannte Eigenschaft. Auf eine gewisse endliche Anzahl ajemanderfolgénder Ziffern 3 folgt also in jeder Kolonne die Ziffer 2.
In ganz derselben Weise geht hervor, dass auf eine gewisse end- liche Anzahl aufeinanderfolgender Ziffern 2 in jeder Kolonne die Ziffer 1 folgt.
Nach dem Bildungsgesetz der betrachteten Kette bleibt ferner — mit Rücksicht auf das Axiom von Archimedes — kein Wert u, v oder w fortwährend der grösste. Auf eine gewisse endliche An- zahl aufeinanderfolgender Ziffern 1 folgt also in jeder Kolonne die Ziffer 2 oder die Ziffer 3.
Aus obigen Erwägungen stellt sich heraus, dass falls wir eine genügend grosse, jedoch endliche Anzahl von aufeinanderfolgenden Zeilen der Tabelle II betrachten, so kommt sowohl die Ziffer 1 als auch die Zijfer 2 in allen drei Kolonnen vor.
Aus diesem ees lasst sich ferner schliessen, dass die Be-
ziehungen : lim P,=lim Q,=lim R, = oo (5) DC MC v=& ‚gelten.
Um die Zahlen P,, Q,, R, sukzessiv aus den Ausgangswerten
y?
Po hee =O
herzuleiten, haben wir nach den Untersuchungen in Nr. 3 auf fol- gende Weise zu verfahren: Der Koeffizient des mittleren von den Werten u,, v,, w, in der
12 ÜBER DEN EUKLIDISCHEN ALGORITHMUS
Gleichung 1=P,u,+Q,v,+R,w, wird durch die Summe von jener selben Zahl und dem Koeffizienten des grössten Wertes u,,
v,, w, ersetzt, während die beiden übrigen von den Zahlen P,,,
>» Q,, R, unverändert gelassen werden, hierbei gehen die Zahlen‘ Py ay Q,+v Ru Es sind mithin P,, Q,, R, ganze Zahlen =O; und von einem gewissen Index v an sind sie tatsächlich positiv. N
Anfangs, die Zahlen P, sind sämtlich positiv, weil P, Vie
Es bezeichne ferner v, die erste Zeile der Tabelle II, welche die Ziffer 1 in der ersten Kolonne aufweist. Dann sind
a) die Zahlen Q, +7 Q, +2 --. sämtlich positiv, falls die Ziffern der Zeile v, in der Reihe 1, 2, 3 aufeinanderfolgen;
b) die Zahlen Ry +v Ry +2 ... sämtlich positiv, falls die Zilfern der Zeile v, in der Reihe 1, 3, 2 aufeinanderfolgen.
Und endlich: im Falle a) sind die Zahlen QL Qu +2 == lich positiv, wenn v, (w,>v,) die erste Zeile angibt, welche die Ziffer 2 in der 3:ten Kolonne aufweist; im Falle b) sind die Zahlen Ry vp R,, 9, sämtlich positiv, wenn v,(¥,>¥,) die erste Zeile angibt, - welche die Ziffer 2 in der 2:ten Kolonne aufweist. Wir haben mithin
Ds Q,; Rye für v>»,.
hervor.
sämt-
Ferner bestimmen wir eine Zahl », derart, dass die Zeilen der Tabelle II vom (,;+1):sten bis zum v,:ten die Ziffer 2 in allen drei Kolonnen aufweisen; dann ist
PQ. Re 2 dünn. Auf dieselbe Weise erhalten wir allgemein
RN ORE Rez fiir v>vnıp
womit die Beziehungen (5) als richtig erwiesen sind.
3 ÜBER DEN EUKLIDISCHEN ALGORITHMUS 13
- Wir setzen nunmehr
d,=P,—-«wP, ru aD, (6) 0, = On o Q,, 0, Q, an 0? Q, 215 =R’—o IR, ; Ab = Ro R, 4
Den grössten von den Werten |d,|, |0,|, | 4,| bezeichnen wir mit M,, den grössten von den Werten |d,|, |d,|, |4,| mit M,, und wir behaupten, dass
(7) | My 1 =M,, My.) =M,.
Nehmen wir wieder einen Augenblick an, dass die Werte LÄR w, der Grösse nach geordnet w,>v,>u, sind, und dass wir als Subtrahenden den Wert v, wählen, um aus dem System S, das System S, ,, herzuleiten. Mit Rücksicht auf die Untersuchungen in Nr. 3 haben wir dann nach (6)
d d, das
v+17 y+1 v a RA Sy oy + Ay A, 14 =A,, A,117 A,
Die Richtigkeit unserer Behauptung geht folglich hervor, falls wir beweisen können, dass
le, | M, é,+4,|<M,.
Es ist nach (2) und (6)
u,d, TV +w,4,=0,
und weil z, +0, erhalten wir daher
UV Ww @ ee), u, U,
14 UBER DEN EUKLIDISCHEN ALGORITHMUS Nunmehr sind zwei Fälle zu unterscheiden:! M008) <a 0 und 2.0.0.1 .0.
Falls 0,4, < 0 ist, haben oe eee (6,1: TA DEU WE Zee Dac Falls aber 0, 4, >0 ist, ergibt sich nach (8)
41-712, [+ —14,1>18,1+14,1=1%, +4], Vv Vv
und wir haben mithin auch in diesem Falle |0, + 4,|<M,, , w.z.b. w.
Auf ganz dieselbe Weise leiten wir die Ungleichung |d, + 4,|<M,, her, und es gelten mithin die Beziehungen (7) unter den oben (vgl. S. 13) gemachten Voraussetzungen betreffend die Werte u, u,,, ,. Falls diese Werte der Grösse nach geordnet in einer anderen Reihe aufeinanderfolgen, oder falls wir als Subtrahenden nicht den mitt- leren sondern den kleinsten von denselben wählen, ergibt sich die Richtigkeit jener Beziehungen in analoger Weise.
Nach (7) haben wir für jeden Index v
d, \ (9) 7
y?
O
y?
ö
4,= M, = max. (1, w) A, = M, = max. (1, 0°).
Vie
Die Beziehungen (9), (6) und (5) zusammengestellt geben den verlangten Konvergenzsatz (4). —
Die in dieser Nummer betrachtete spezielle Kette wurde von
Viggo Brun (En generalisation av kjedebroken I, II; Videnskaps- selskapets Skrifter, Kristiania, 1919, 1920) angegeben. Auf geo-
metrischem Wege zeigt Brun u. A., dass wenigstens eine von den
Quoten P:P,, Q,:Q,, Ry:R, mit wachsendem Index wi gegen 0 KONG nd dass ebenso wenigstens eine von den Quoten a OD SG ITIL wachsendem Index v gegen w? kon- vergiert. Abo, den 28. Januar 1922.
Es ist J, 4, 0, weil die betrachtete Kette unserer Annahme gemäss nicht abbricht, und w mithin keine rationale Zahl ist. (Vel. S. 8.)
N
VS
iv
Fe
ER
PRIS 45 FMK
REDAKTIONENS ADRESS: ÅBO AKADEMI, ÅBO, FINLAND
. 4a- AAA A age .behhAbbi POPE EIT i Fra ee a an Ja BITIT ALS OV I a AAA Ie =a WY VVY Lolli Gal pees an Rar sa |||
sonnel rina agen | ppanAgsfnall ON hl
A A. ‘mA ~ a 3 er eh im & > a IL een Tt a ahnen, Zar iat nial I —on@mka Smt Å auna-al ae Napoht ces malls ex! Se Pee eet tester LoL Li ry
u | | um. 111) Sani: ae res - rs TEE un TAU HINNA | | RR, rr I || INT TYG ä HT SIT A an. Ir Ty AA an Ann N TA | fikon Snel SUPA tae ae | phe | | Var m 5 san INT Za? aft | | | N, | a. x Y d SRA AA casera nso pee SRR ‘ Wplaaaan bh ! en | a } 3 MALA MAR re LETAR RvR I IT Peay ÜBER AN anna |
IE NY | yy 1 ML TR Ay, Aan BF ANS gh scaler agit? Ty rerr aand RUN ge (pawesah an An An
aan NN PAR RADAR em iii N NAVET AT YIN Bade...
m 2 7%
Nr
„=. d PA
As PTETLILEE ER | SAAANVAKRA
RER =
SARA 7 mtr N | lå By UNASAARAP I Y :
RP ”
ar : ll FANAKARAA GAA enn | LL
tT nin HTT my yr
y
Ran stin gi? BE
‘Va, 6
sa BASEL Ag
FRV WS, VA
: 2A.
ms Aj . a A nl > i SUN NA AARAAPBPA a PAA i ) VW \ | Auen a Ba: Hoch Ra 2 AN UN, (Ne MAN NN A J SI A NEN AA IR VN VA 4 Aa A
a A | FAT SURA: RES A | IR heen N aw
Ps. | 2 „rARAARLAA N, a A Try TILLIT TVIVEL TT a WT AR RAAAD SAA A m ana = =. Se RER
„ A A, DR TY TY FRA
HT test NRA i iain” : | Ai N RR ON nn AA ÄNNNNA RAN ARA AN 7 ARM ; & may ee j yo an Ran ra aan RAMA aan
oregano BMOENFENR „urn ARA I LaELIERDAE ÖRA ÄLG wea so a) Dib OS % mM NAR RM > t | " AReBAesab, | rd tt
MR nt
SA m 4 - eae (327 pp, ÅA "Mae bt ww h > Wy, es Lam hen! HU let s u a Th Ve a;
} IN WE in rm ADP aN OP, x
ns 2 SÅ N Ra ana “gas; ve: WARS ak un. NAN an,
th ant doh es ara SE r \
| yey, eee Tih 2 en MA ae! Aa
a Hal be. hl ee a
ee Ren WO ja? Ex art u ARN Nid ad Marl Aa FEN da 7494 ae 17 R NTA föra. yt Jab va PR AU TY | LEZSESBESZDN Vy ma Wr ‘a
>| Aneır st4e a" PP x ae afin no in, & af (3 KAAARAAEN Axi.
a
Alay, Ta as | Bl agin’, SER DOT Tia lena ln
q : RE : ‘ wala. MOMMUEL DERE T T/DERSEAMMAGAReMane Na rege x “ape 1 I « \ N EN en bi al) NY YR
=. 4 Awe as | a, ap’ a ‚x . > ;
BEE IH da eames A Vy Verviers sheen Ret Altersscass oo. IH) Aa pod = = sö : Pr RE
u. *: > 35 Ib A wai! AAR an de ape awl nce eas : SMAL. |
Tr 7 ER ‘a An mite ~~ ee AN ara, we ae Mat) | un‘ på Pp ap
2 SITE ine, REN Tre Br sberpig ater DEA var Hil Niue DAR ana ARV IN Ye Ase yea Yan.
he tron NMG! FEI h) An “
i I 6 a PAY aan deu - = ..2 ae 7 R
BE öra LT WR pA Th MYR Kalas gg, NASA AN haat
MH 40944! ar SENS AN
MAT Vu IIS | fh ny | | ta AC in x
| urn Ba cael . Ay, „ART | SNe TATE] a TT YY SARA ‚aA 4 one LD suis > er Weyer, zw || WWW LV SAA La, a | NAN Ye
va. HARRI miata ab, LER sy
{ a der { u AR sr Lane | AR 4,4 | |
rar a UN AN Haha un” ae /N& RE Ipod 2 se x” BERN m aap UUAA ; ate ree bb} IMa
Ne, ob ti. AT LJ fon ip’ a -® 28094 2 30 * TN Zar VAR an th un’ > 0 52 MAR AN, =| Db lye A ae Leh böj Raw 2 I Avi I tä IL: PET Fy j ae
vi | HATT vr nut mei INA RS ale AN FELGE mA” n den | js Masten
| r j PA 1 § N u ar a Få SEMEL) o> he bo BAY VAN > 4 I a > Uhl... Bee ven: BR m ey ee ae Big MRE GRD m ers N27 ¢ Tt pe a he Se Za Mins dh my FT ee PD Abe
uli
3 9088 01